Compare commits

...
Author SHA1 Message Date
Finesssee 5058240026 feat: add Rust account state import export 2026-05-18 20:13:05 +07:00
Finesssee 44b6e577ec feat: add Rust Codex route parity 2026-05-18 20:08:26 +07:00
Finesssee d846b2a56f feat: make Rust TUI provider-aware 2026-05-18 20:02:32 +07:00
Finesssee cc9d1baf51 feat: add Rust management status endpoints 2026-05-18 19:57:51 +07:00
Finesssee 313d86d7d5 feat: add Claude messages route 2026-05-18 19:54:12 +07:00
Finesssee 76e32b6032 test: harden Codex auth import 2026-05-18 19:48:31 +07:00
Finesssee f62e779d86 test: add Rust Codex smoke check 2026-05-18 19:44:25 +07:00
Finesssee 25b7aa62e3 feat: start Rust replatform branch 2026-05-18 19:36:43 +07:00
Finesssee a1ce124748 Document OAuth account risk 2026-05-18 19:16:51 +07:00
Finesssee cedb38d1cf Harden management and OAuth surfaces 2026-05-18 19:10:42 +07:00
RCDandCopilot c36e214e2e fix: remove duplicate TUI flag registration
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-05-18 11:54:34 +00:00
Finesssee c26de75913 Harden PR and release workflows 2026-05-12 18:42:12 +07:00
RCD bd3901ca29 chore: merge upstream CLIProxyAPI v7.0.2
# Conflicts:
#	internal/translator/openai/openai/responses/openai_openai-responses_request_test.go
2026-05-11 07:53:17 +00:00
RCDandCopilot 3636be31d6 fix: make Kiro auth fallback safe and document Claude limits
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-05-11 07:51:59 +00:00
Luis Pater 1fca942b9c Merge remote-tracking branch 'origin/pr/3239' 2026-05-10 20:43:35 +08:00
RCDandCopilot 18d9369ce4 chore: merge upstream CLIProxyAPI v7.0.1
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-05-10 07:47:19 +00:00
Luis Pater 5e5b1bce35 feat(config): add detailed logging for home config changes
- Introduced `logHomeConfigChanges` to compare old and new configs, logging detected differences.
- Leveraged `diff.BuildConfigChangeDetails` for structured change detection.
- Adjusted logging behavior to enable debug-level logs dynamically when required.
2026-05-10 15:28:49 +08:00
Luis Pater 15ac7fb932 refactor(auth): simplify home auth session management and remove ref counting
- Consolidated `homeRuntimeAuths` to store a map of session-scoped auth maps, replacing `homeRuntimeAuthSessions` and `homeRuntimeAuthRefs`.
- Adjusted session cleanup logic to directly remove session-scoped auths without reference counting.
- Added `GetExecutionSessionAuthByID` to retrieve auths scoped to a specific execution session.
- Updated tests to reflect the new session-scoped caching behavior.
2026-05-10 15:21:33 +08:00
Luis Pater 8300ee8bbe feat(auth): enhance home auth session reuse with scoped caching and ref counting
- Added `homeRuntimeAuthSessions` and `homeRuntimeAuthRefs` for scoped caching of home auths per session.
- Updated `pickNextViaHome` to prevent reuse of already-tried pinned auths during session retries.
- Implemented reference counting for shared auths across multiple sessions to improve memory management.
- Enhanced session cleanup logic to clear cached auths only when all referencing sessions are closed.
- Added unit tests to validate scoped caching, retry logic, and session cleanup behavior.
2026-05-10 14:00:13 +08:00
Luis Pater dc1cc7f115 feat(auth): add websocket session reuse for home auths with caching support
- Introduced `homeRuntimeAuths` to cache home auths for websocket session reuse.
- Updated `pickNextViaHome` to prioritize cached auths for pinned websocket sessions.
- Implemented automatic clearing of cached home auths when home mode is disabled.
- Added unit tests to validate caching behavior, clearing logic, and fallback scenarios.
2026-05-10 13:39:14 +08:00
Luis Pater a44e5eb1ab Merge branch 'v7' into dev 2026-05-10 02:33:42 +08:00
Luis Pater 67fb4eb98e feat(auth): add shouldReturnLastErrorOnPickFailure helper and improve error handling in home mode
- Introduced `shouldReturnLastErrorOnPickFailure` to streamline error return logic during provider selection.
- Added `isHomeRequestRetryExceededError` for better home-specific error classification.
- Updated fallback conditions to enhance error handling clarity in `pickNextMixed`.
2026-05-10 02:09:53 +08:00
Luis Pater 66c3dae06b feat(home): implement count for home auth dispatch requests and enable usage statistics
- Added `count` attribute to `homeAuthCount` requests to improve home message batching.
- Enabled usage statistics for home mode by default and added config-level enforcement.
- Adjusted failure logging to include detailed metadata in `UsageReporter`.
- Updated multiple executors to pass error details to `PublishFailure` for better debugging.
- Enhanced unit tests to validate `count` behavior and usage statistics enforcement across components.
2026-05-10 01:30:43 +08:00
Luis Pater 1abf8625d8 feat(logging): add home request-log forwarding support
- Introduced `SetHomeEnabled` to enable/disable request-log forwarding to the home control plane.
- Implemented `forwardRequestLogToHome` for non-streaming logs and `homeStreamingLogWriter` for real-time streaming logs.
- Enhanced `FileRequestLogger` to bypass local logging when home forwarding is enabled.
- Updated server configuration to dynamically toggle home request-log forwarding based on changes.
- Added corresponding unit tests to ensure correct forwarding behavior and fallback mechanisms.
2026-05-09 23:39:59 +08:00
Luis Pater 3cbd587b2c Merge pull request #3283 from wuchulonly/fix/responses-ws-tool-output-context
Fix Responses WebSocket tool output context repair
2026-05-09 21:08:43 +08:00
Luis Pater 41f4ee7c7d feat(auth): enhance auth index generation with improved file path handling
- Updated `EnsureIndex` logic to incorporate absolute and cleaned file paths when generating auth indexes.
- Refined metadata handling to include OAuth type in auth index seed.
- Improved compatibility for `json` file paths as sources in auth attributes.
- Added unit tests to validate correct auth index behavior for various path and type scenarios.
2026-05-09 21:03:11 +08:00
Luis Pater c69ff49758 feat(auth): add support for persisting disabled flag in token storage
- Updated `FileTokenStore` and related stores (`objectstore`, `gitstore`, `postgresstore`) to include the `disabled` flag in metadata for token storage.
- Adjusted `Auth` metadata handling to initialize empty maps when absent.
- Refined logic in `auto_refresh_loop` and `conductor` to exclude `disabled` tokens from refresh checks.
- Added comprehensive unit tests to verify proper handling of the `disabled` flag in storage and retrieval operations.
2026-05-09 19:48:42 +08:00
Luis Pater 68fddaa319 Merge pull request #3292 from lihan3238/fix-3272
fix: apply default auth-dir when config value is empty
2026-05-09 17:34:05 +08:00
Luis Pater 09ac8a1165 Merge pull request #3221 from mochenya/main
fix(executor): ignore null OpenAI stream usage chunks
2026-05-09 11:53:26 +08:00
Luis Pater 0dcb8bd714 refactor(cliproxy): remove ClaudeCodeSessionAffinity support and simplify session affinity logic 2026-05-09 10:51:49 +08:00
Luis Pater 0f0fcd2304 feat(config): add per-auth disable_cooling override support
- Introduced `disable_cooling` metadata field for fine-grained control over cooldown scheduling.
- Updated `Auth` object to include `Metadata` with conditional logic for handling empty states.
- Added YAML configuration support for `disable_cooling` in API key definitions across providers.
- Enhanced unit tests to validate `disable_cooling` behavior in various scenarios.
2026-05-09 10:51:27 +08:00
Luis Pater c67096b687 feat(server): add support for loading configuration from a remote home control plane
- Introduced `-home` and `-home-password` flags for specifying home control plane address and authentication.
- Implemented fetching and parsing configuration from the home control plane when `-home` is used.
- Adjusted server configuration handling to bypass local config files when loading from home.
- Ensured compatibility with cloud deploy mode and validation of home configurations.
2026-05-09 07:14:44 +08:00
Luis Pater 1721994111 feat(management): expose additional OAuth and configuration helpers
- Added new helper methods for OAuth session management (`RegisterOAuthSession`, `CompleteOAuthSession`, etc.).
- Introduced `WriteConfig` for persisting management configurations.
- Exported `Handler` type and `NewHandler` constructors for SDK consumers.
2026-05-09 00:23:45 +08:00
lihan3238 4cbe172934 refactor: extract DefaultAuthDir constant per review feedback 2026-05-08 22:28:38 +08:00
lihan3238 4071fdef84 fix: apply default auth-dir when config value is empty
When auth-dir is not specified in config.yaml, ResolveAuthDir returns
an empty string which causes os.MkdirAll to fail with no path. Use
the documented default ~/.cli-proxy-api instead.

Fixes #3272
2026-05-08 21:47:41 +08:00
Codex c883114a4d fix responses websocket tool output context 2026-05-08 05:12:30 +00:00
Luis Pater e50cabac4b chore: upgrade CLIProxyAPI dependency to v7 across the project
- Updated all references from v6 to v7 for `github.com/router-for-me/CLIProxyAPI`.
- Ensured consistency in imports within core libraries, tests, and integration tests.
- Added missing tests for new features in Redis Protocol integration.
2026-05-08 11:46:46 +08:00
Luis Pater 785b00c312 Merge pull request #3237 from seakee/docs/add-cpa-manager-usage-statistics
docs: add CPA-Manager to usage statistics recommendations
2026-05-07 09:32:32 +08:00
Luis Pater a034cf8b8d Merge pull request #3247 from edlsh/fix/amp-thread-actors-route
fix(amp): proxy thread actors route
2026-05-07 09:31:16 +08:00
edlsh 01171742a6 fix(amp): proxy thread actors route 2026-05-06 13:12:35 -04:00
Luis Pater fb08b92402 feat(executor): add upstream disconnect handling for Codex WebSocket sessions
- Introduced `UpstreamDisconnectChan` for Codex WebSocket sessions to notify downstream connections of upstream disconnections.
- Implemented `notifyUpstreamDisconnect` to signal errors and close channels on disconnect events.
- Added integration tests to validate WebSocket session behavior on upstream disconnect.
- Updated OpenAI WebSocket response handlers to properly close connections upon upstream disconnect notifications.
2026-05-06 22:09:33 +08:00
AhDEV a09a16e8f6 Merge branch 'router-for-me:main' into my-fix 2026-05-06 16:41:14 +08:00
RCDandCopilot 68b97075af chore: merge upstream CLIProxyAPI v6.10.8
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-05-06 08:10:59 +00:00
seakee ad3f4f2ce5 📝 docs(readme): add CPA-Manager usage statistics recommendation
Add CPA-Manager to the Usage Statistics recommendations across English, Chinese, and Japanese READMEs.

Highlight request-level monitoring, cost estimation, LiteLLM price sync, SQLite persistence, and Codex account-pool operations for multi-account maintenance.
2026-05-06 15:49:57 +08:00
songyu 17a1f53c47 fix:openai 2 kimi error Continuous function_call
连续的function_call 转换 tool_calls
2026-05-06 14:37:18 +08:00
Luis Pater ed1458aa6d chore(docs): update sponsor details in README
- Replaced sponsor `z.ai` with `PackyCode` and updated related descriptions, images, and links in `README.md`, `README_CN.md`, and `README_JA.md`.
- Removed outdated sponsor entries for `Poixe AI` in all README files.
- Added new image assets for PackyCode (`packycode-cn.png` and `packycode-en.png`).
2026-05-06 00:41:50 +08:00
mochenya 99dfbaef61 fix(executor): ignore null OpenAI stream usage chunks
- Added validation so OpenAI-style usage parsing only accepts object payloads with token fields.
- Prevented streaming usage:null chunks from publishing zero-token records before the final usage chunk arrives.
- Reused the shared OpenAI-style parser for stream usage to support both chat completions and responses token field names.
- Added tests covering null usage chunks and input/output token usage fields in streaming responses.
2026-05-05 12:31:33 +08:00
Luis Pater da6c599efd refactor(management): rename GetUsage to GetUsageQueue and update routes/tests
- Renamed handler and test methods for better clarity on functionality.
- Updated route from `/v0/management/usage` to `/v0/management/usage-queue`.
- Adjusted integration and unit tests to reflect new naming and routes.
2026-05-05 03:02:25 +08:00
Luis Pater 61b39d49bd feat(management): add usage record retrieval endpoint
- Implemented `/v0/management/usage` endpoint for fetching queued usage records from Redis.
- Included validation for `count` parameter to ensure positive integers.
- Added unit tests for queue retrieval and validation, with authentication validation in integration tests.
- Updated management routing to include the new endpoint.
2026-05-05 02:53:04 +08:00
Luis Pater ba5d8ca733 feat(usage): add support for requested model alias handling
- Introduced methods for setting and retrieving model aliases in execution and usage contexts.
- Enhanced `UsageReporter` and related structures to include client-requested aliases.
- Updated tests to validate alias propagation and ensure correct usage reporting.
- Adjusted metadata handling in CLIProxyAPI executors to address alias integration.
2026-05-05 01:47:53 +08:00
Luis Pater 28b4b19e7e Merge pull request #3208 from kdcokenny/codex-websocket-protocol-parity
Align Codex websocket protocol semantics
2026-05-05 01:29:19 +08:00
Luis Pater bdc424007e Merge pull request #2896 from edlsh/fix/oauth-tool-rename-per-request-map
fix(amp): smart-mode tool name fixes + deep-mode response repair
2026-05-05 00:58:39 +08:00
Luis Pater e4a93c02c5 fix(executor): enhance parsing of OpenAI stream data lines
- Added trimming for stream input lines to prevent processing of unnecessary whitespace.
- Improved handling of unsupported prefixes and malformed JSON responses, ensuring errors are recorded and propagated appropriately.

Fixed: #2690
2026-05-04 23:42:26 +08:00
Luis Pater 8262a03f29 Merge PR #2568: fix Claude refresh backoff 2026-05-04 21:44:11 +08:00
Luis Pater ecf1c2590c fix: preserve Antigravity cancellation errors 2026-05-04 21:18:18 +08:00
Luis Pater 162897e02a Merge remote-tracking branch 'origin/pr/3205' into dev 2026-05-04 21:17:01 +08:00
Luis Pater c1caa454b3 fix(translator): handle empty tool function names in OpenAI Claude responses
- Added check to prevent processing of empty `function.name` values, ensuring valid data is handled.

Fixed: #2557
2026-05-04 21:00:33 +08:00
Luis Pater bf6fa402e2 fix(executor): strip Vertex OpenAI response tool call IDs for consistency
- Integrated `StripVertexOpenAIResponsesToolCallIDs` to remove tool call ID data from request bodies and translated requests.
- Ensures uniformity and avoids unnecessary payload data propagation.

Fixed: #2549
2026-05-04 17:54:16 +08:00
Luis Pater 85c0150653 feat(translator): add token usage tracking and improve usage handling
- Introduced `claudeUsageTokens` struct for detailed token usage tracking.
- Replaced `calculateClaudeUsageTokens` with `Merge` and `OpenAIUsage` methods for better modularity.
- Enhanced integration of usage tokens into response processing, enabling more accurate reporting of token details.

Fixed: #2419
2026-05-04 16:57:50 +08:00
Luis Pater 89d80bfff4 fix(executor): adjust ApplyThinking order and add payload override test
- Moved `ApplyThinking` logic earlier in `openai_compat_executor` to align with configuration application sequence.
- Added test to verify payload override precedence over Thinking suffix configuration.
2026-05-04 16:45:25 +08:00
Luis Pater a1eba112f3 Merge pull request #2416 from kslamph/fix/gemini-cli-projectid
fix(gemini-cli): use backend project ID from onboarding response
2026-05-04 16:08:31 +08:00
Kenny 6b4bc0a9a8 Align Codex default identity and docs 2026-05-03 21:13:37 -07:00
Kenny 08b0fe6816 Fix Codex websocket retry metadata 2026-05-03 19:01:44 -07:00
Kenny c19ae1d5be Align Codex websocket protocol semantics 2026-05-03 15:56:39 -07:00
Luis Pater 17be6442a8 fix(translator): improve tool response handling for non-string content
- Added `setToolCallOutputContent` to process various content types, including arrays and fallback cases.
- Implemented robust handling for specific tool output types like text, image URLs, and files, ensuring proper serialization.
- Improved fallback logic to handle unexpected or missing data.

Fixed: #2313
Closes: #2349
2026-05-04 05:50:01 +08:00
Luis Pater 38dad2afdf chore(docker): upgrade base image to alpine 3.23
Fixed: #2265
2026-05-04 05:36:09 +08:00
Luis Pater 8e6ef3fa64 fix(websocket): ensure state consistency on auth errors in streaming
- Added logic to reset `pinnedAuthID` and replay transcript on unauthorized, forbidden, or throttling errors.
- Enhanced error handling in `forwardResponsesWebsocket` with detailed status inspection.
- Introduced `shouldReleaseResponsesWebsocketPinnedAuth` to determine auth reset conditions.
- Updated state management to preserve prior request and response data during forced replay.

Fixed: #2230
2026-05-04 05:23:23 +08:00
Luis Pater a1487b0958 fix(translator): handle non-string types in tools result processing
- Skip setting values for non-string `type` fields to prevent runtime errors.

Closes: #2226
2026-05-04 05:08:31 +08:00
Luis Pater 82ebe24b9e Merge pull request #2266 from DragonFSKY/fix/ws-compact-tool-output-mismatch
fix(websocket): skip stale state merge after client-side compact
2026-05-04 04:40:43 +08:00
Luis Pater 2753d9fb71 feat: add validation for Claude streaming responses
- Implemented `validateClaudeStreamingResponse` to ensure upstream streaming data integrity.
- Added new tests to verify response validation, including empty streams, error events, incomplete streams, and valid streams.
- Integrated validation logic into the Claude executor's streaming handler, returning detailed errors for malformed upstream data.

Fixed: #2193
2026-05-04 03:37:31 +08:00
1137043480 bf0e5c23f7 fix: prevent goroutine leaks in streaming executors via context-aware channel sends
All streaming executors use bare channel sends (out <- chunk) inside goroutines
that process upstream SSE responses. When the downstream consumer disconnects
(client timeout, network drop, etc.), these sends block indefinitely, causing
the goroutine and all associated resources (HTTP response body, scanner buffers,
translation state) to leak permanently.

Over time, leaked goroutines accumulate monotonically, leading to RSS growth
from ~30MB to 3.7GB+ and eventual OOM kills on resource-constrained VPS hosts.

Fix: Replace all bare 'out <- ...' sends with:
  select {
  case out <- ...:
  case <-ctx.Done():
    return
  }

This ensures goroutines terminate promptly when the request context is canceled,
allowing GC to reclaim all associated resources.

Affected executors (9 files, 36+ send sites):
- antigravity_executor.go (5 sites)
- gemini_cli_executor.go (6 sites)
- gemini_vertex_executor.go (6 sites)
- aistudio_executor.go (4 sites)
- gemini_executor.go (3 sites)
- openai_compat_executor.go (3 sites)
- claude_executor.go (4 sites)
- codex_executor.go (2 sites)
- kimi_executor.go (3 sites)
2026-05-03 11:25:04 -04:00
Luis Pater 672fdd14ed feat: filter and drop empty assistant messages in Kimi executor
- Added `filterKimiEmptyAssistantMessages` to identify and remove empty assistant messages with no content, tool links, or reasoning.
- Integrated logging to track the number of dropped messages.
- Updated tests to validate the filtering logic for both empty and valid assistant messages.

Fixed: #1730
2026-05-03 22:40:42 +08:00
Luis Pater af65908cb0 feat: enhance tool mapping with namespace and web search support
- Added functions to handle tool conversion, including namespace-based tools and web search tools.
- Improved parameter normalization and tool input schema standardization.
- Integrated logic to handle qualified tool names and map override functionality.
- Refactored existing tool processing for better extensibility and maintainability.

Fixed: #3199
2026-05-03 22:26:23 +08:00
Luis Pater 756b600b7a Merge pull request #3204 from zhanglunet/add-usage-dashboard-link
Add CLIProxyAPI Usage Dashboard to usage statistics docs
2026-05-03 21:38:45 +08:00
Luis Pater 054d0dee1d Merge pull request #3201 from murasame612/docs/add-cliproxy-pool-watch
docs: add CLIProxy Pool Watch project
2026-05-03 21:37:36 +08:00
zhangluandgemini-code-assist[bot] d2386a3114 Update README_JA.md
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-05-03 20:38:51 +08:00
zhangluandgemini-code-assist[bot] 7972130513 Update README_CN.md
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-05-03 20:38:25 +08:00
John 81db7fdc1e Add CLIProxyAPI Usage Dashboard to statistics docs 2026-05-03 20:23:23 +08:00
ziwu 5fc6f662e1 docs: add CLIProxy Pool Watch project 2026-05-03 18:25:11 +08:00
Luis Pater 593995a404 Merge pull request #3197 from VjayC/dev
docs: update Subtitle Translator project description
2026-05-03 14:23:44 +08:00
Vijay Chimmi 101b59cfe8 docs: update Subtitle Translator project description 2026-05-02 17:37:38 -07:00
Luis Pater 56df36895a feat: add configurable retention period for Redis usage queue
- Introduced `redis-usage-queue-retention-seconds` config parameter with a default of 60 seconds and a max of 3600 seconds.
- Updated logic in `redisqueue` to honor configurable retention periods for enqueued usage data.
- Modified config validation and initialization to support and enforce retention limits.
- Enhanced change tracking in `config_diff` to detect updates to this parameter.
2026-05-02 20:43:16 +08:00
Luis Pater 85124f098b Merge pull request #3194 from daishuge/codex/add-daishuge-cpa-port-20260502
docs: add Playful Proxy API Panel
2026-05-02 17:43:02 +08:00
daishuge 2efa56dbb8 docs: add Playful Proxy API Panel 2026-05-02 17:03:11 +08:00
Luis Pater 79579c34bf docs: update README to consolidate and clarify CPA Usage Keeper details
- Moved CPA Usage Keeper from "CLI tools" to a dedicated "Usage Statistics" section.
- Added details on its functionality, periodic data sync, SQLite storage, and built-in dashboard.
- Applied updates across English, Chinese, and Japanese README files for consistency.
2026-05-02 13:37:42 +08:00
Luis Pater 18bb9c315f chore: remove usage tracking and logging functionality
- Deleted the `LoggerPlugin` along with associated usage tracking and in-memory statistics logic.
- Removed all related tests (`logger_plugin_test.go`, `usage_tab_test.go`) and external-facing handler (`usage.go`) for usage statistics export/import.
- Cleaned up TUI integration by deleting `usage_tab.go`.
2026-05-02 04:50:58 +08:00
Luis Pater b8bba053fc feat: add tracking for auth request success and failure counts
- Introduced `Success` and `Failed` fields in auth records to track request outcomes.
- Updated `/v0/management/auth-files` and `/v0/management/api-key-usage` responses to include success and failure counts.
- Enhanced tests to validate tracking logic and API responses.
2026-05-02 03:40:00 +08:00
Luis Pater 8c2f1a80d3 feat: enhance API key usage grouping with base URL inclusion
- Updated `GetAPIKeyUsage` to group API key usage by "base_url|api_key" composite keys.
- Adjusted logic to handle `base_url` extraction from auth attributes.
- Revised unit tests to validate "base_url|api_key" grouping behavior.
2026-05-02 02:20:49 +08:00
Luis Pater e37f3be0bf chore: update .goreleaser.yml to include custom archive naming with arch override logic 2026-05-02 00:09:08 +08:00
Luis Pater b0dc9df887 feat: add API key usage endpoint with provider and key grouping
- Implemented `GetAPIKeyUsage` to expose recent request data grouped by provider and API key.
- Added supporting function `mergeRecentRequestBuckets` for bucket aggregation.
- Registered new endpoint `/v0/management/api-key-usage` in the management API.
- Included extensive unit tests for provider and key-based grouping validation.
- Updated `formatRecentRequestBucketLabel` to support configurable bucket duration.
2026-05-01 23:34:18 +08:00
Luis Pater 6187919000 feat: add support for recent request tracking in auth records
- Implemented `RecentRequestsSnapshot` in `Auth` to capture bucketed recent request data.
- Added new fields and methods to `Auth` for tracking request success and failure counts over time.
- Updated `/v0/management/auth-files` response to include recent request data for each auth record.
- Introduced unit tests to validate request tracking and snapshot generation logic.
2026-05-01 22:55:22 +08:00
Luis Pater 4035abc0cd refactor(logging): replace gin-specific context handling with generic context-based request metadata utilities
- Introduced reusable utilities in `requestmeta` to manage endpoint and response status in request contexts.
- Refactored plugins and handlers to use context-based metadata, removing direct dependency on `gin`.
- Updated tests to validate new context utilities and replaced `gin`-based context handling.

Fixed: #3166
2026-04-30 23:36:07 +08:00
RCDandCopilot 101099ca3a chore: merge upstream CLIProxyAPI v6.9.45
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-04-30 13:35:28 +00:00
Luis Pater 8b286e8fb3 Merge pull request #3163 from C4AL/docs/add-codexclipplus-ecosystem
docs: add CodexCliPlus to README ecosystem list
2026-04-30 20:42:32 +08:00
C4AL aa70d13f60 docs: add CodexCliPlus to README ecosystem list 2026-04-30 20:36:37 +08:00
songyu 05ecfb6241 feat: add local Docker build script and update compose configuration
- Introduced a new script `docker-build-local.sh` to build a local Docker image and start services using Docker Compose.
- Updated `docker-compose.yml` to allow dynamic pull policy configuration via the `CLI_PROXY_PULL_POLICY` environment variable.
- Modified `Dockerfile` to support build arguments for Go module proxy settings during the `go mod download` step.
2026-04-30 14:01:56 +08:00
songyu 243c582159 feat: add unit tests for OpenAI responses request conversion
- Introduced a new test file for validating the conversion of OpenAI responses to chat completions.
- Implemented tests to ensure correct merging of consecutive function calls and proper handling of interrupted function calls.
- Enhanced the main conversion function to buffer consecutive function calls and emit them as a single assistant message.
2026-04-30 13:33:40 +08:00
Luis Pater 6ba7c810a7 feat: apply image_generation filtering before payload rules
- Updated `ApplyPayloadConfigWithRoot` to prioritize `disable-image-generation` filtering before applying payload rules.
- Ensured payload overrides can explicitly re-enable `image_generation` when required.
- Added unit tests to validate `image_generation` restoration through overrides.
2026-04-30 12:42:08 +08:00
Luis Pater f56a19e5b8 feat: add tri-state support for disable-image-generation configuration
- Introduced `DisableImageGenerationMode` with support for `false`, `true`, and `chat` values.
- Updated payload handling to preserve `image_generation` on images endpoints when `chat` mode is enabled.
- Modified OpenAI image handlers (`ImagesGenerations`, `ImagesEdits`) to respect tri-state logic.
- Added unit tests for `DisableImageGenerationMode` behavior and endpoint-specific handling.
- Enhanced configuration diff logging to support `DisableImageGenerationMode`.
2026-04-30 12:10:27 +08:00
Luis Pater 46018417ad feat: remove tool_choice for image_generation when disabled
- Added logic to remove `tool_choice` entries of type `image_generation` from payloads when `disable-image-generation` is enabled.
- Updated `ApplyPayloadConfigWithRoot` to handle new removal logic.
- Added unit tests to verify `tool_choice` removal behavior.
2026-04-30 08:24:14 +08:00
RCDandCopilot dbf432c3e7 chore: merge upstream CLIProxyAPI v6.9.43
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-04-29 23:41:52 +00:00
Luis Pater e3e60f914b feat: support disabling image generation globally
- Added `disable-image-generation` configuration flag to disable the `image_generation` tool globally.
- Updated payload handling to remove `image_generation` tools from request payload arrays when the flag is enabled.
- Modified OpenAI image handlers (`ImagesGenerations`, `ImagesEdits`) to return 404 when the feature is disabled.
- Enhanced configuration diff logging to track changes for the `disable-image-generation` flag.
- Added accompanying unit tests for the new feature in payload helpers and image handler logic.
2026-04-30 03:42:27 +08:00
RCDandCopilot 270e273254 chore: merge upstream CLIProxyAPI v6.9.42
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-04-29 15:38:08 +00:00
Luis Pater 359ec30d0c chore(docs): remove LingtrueAPI sponsorship section from README files 2026-04-29 23:13:12 +08:00
Luis Pater a1f0ed9575 Merge pull request #3071 from sususu98/fix/antigravity-credits-log
Mark Antigravity credits requests in access logs
2026-04-29 22:56:41 +08:00
Luis Pater b3b92f334e Merge pull request #3125 from sususu98/fix/gemini-cli-usage-metadata-variants
fix: parse gemini cli usage metadata variants
2026-04-29 22:56:14 +08:00
Luis Pater 2c1539ead7 Merge pull request #3133 from sususu98/optimize/claude-codex-translator
fix: align claude codex translation
2026-04-29 22:55:37 +08:00
sususu98 0d107dd566 fix: respect declared claude web search tool names 2026-04-29 19:24:53 +08:00
sususu98 1c0c426b85 fix: align claude codex translation 2026-04-29 18:47:03 +08:00
sususu98 4982512da2 fix: parse gemini cli usage metadata variants 2026-04-29 13:10:53 +08:00
Luis Pater 2ea8f77efb feat(models): add GPT-5.5 to the registry with support for advanced tasks 2026-04-29 09:49:26 +08:00
Luis Pater f95ab4cdf1 Merge pull request #3089 from XYenon/feat/session-affinity
feat: support Codex/PI session headers for session affinity
2026-04-28 23:54:16 +08:00
sususu98 c4965befe7 Merge pull request #3117 from sususu98/fix/restore-antigravity-ua-handling
fix: restore Antigravity user agent handling
2026-04-28 19:07:05 +08:00
sususu98 0e1235122e fix antigravity client agent headers 2026-04-28 19:04:40 +08:00
sususu98 e78d45acc9 fix antigravity user agent handling 2026-04-28 19:04:40 +08:00
sususu98 b34f3be13e Merge pull request #3114 from Xbang0222/fix/antigravity-credits-ua
fix(antigravity): 使用真实 antigravity UA 查询积分余额
2026-04-28 19:03:21 +08:00
Luis Pater 9fb6a49260 test(api): add validation for unsupported models in OpenAI image handlers
- Introduced tests to ensure unsupported models are rejected in `/images/generations` and `/images/edits`.
- Added `isSupportedImagesModel` and `rejectUnsupportedImagesModel` functions for consistent model validation.
- Enhanced image handler logic to apply validation checks for model compatibility.
2026-04-28 17:19:12 +08:00
xbang a992dee4e8 fix(antigravity): use real antigravity UA when polling credits balance
The loadCodeAssist polling call hardcoded the User-Agent to
google-api-nodejs-client/9.15.1. Google Cloud Code returns the
paidTier object WITHOUT the availableCredits array for that UA,
so updateAntigravityCreditsBalance always saw "no credits", set the
hint to Available=false for every Google One AI Ultra account, and
the conductor-level credits fallback could never find a candidate.

Switching to resolveUserAgent(auth) (the same UA used for
streamGenerateContent / generateContent) makes the response include
availableCredits, so the credits hint is populated correctly and the
fallback can actually inject enabledCreditTypes:["GOOGLE_ONE_AI"]
when free tier is exhausted.
2026-04-28 16:21:15 +08:00
XYenonandAmp 3ac39dcc7d feat: support Codex/PI session headers for session affinity
Amp-Thread-ID: https://ampcode.com/threads/T-019dce25-c070-773a-ac52-11c541220b30
Co-authored-by: Amp <amp@ampcode.com>
2026-04-27 17:10:50 +08:00
Luis Pater 34027da7f1 Merge pull request #3088 from sususu98/codex-claude-reasoning-signature
Preserve Codex reasoning signatures for Claude
2026-04-27 16:59:11 +08:00
sususu c523101439 Preserve Codex reasoning signatures for Claude 2026-04-27 16:51:35 +08:00
Luis Pater d85e13b044 fix(codex): include content field in reasoning item initialization 2026-04-27 16:41:23 +08:00
Luis Pater 01e16a8509 feat(codex): handle thinking-signature conversion for reasoning content
- Implemented `appendReasoningContent` to support processing of `thinking` signature and text as reasoning input.
- Added test cases to validate reasoning content conversion with and without text.
2026-04-27 16:31:26 +08:00
Luis Pater 04a336f7df fix(usage_helpers): skip zero-token usage in additional model records
- Added `buildAdditionalModelRecord` to filter out zero-token usage details.
- Introduced `hasNonZeroTokenUsage` helper function for token usage validation.
- Updated tests to cover scenarios for zero and non-zero token usage.
2026-04-27 10:56:22 +08:00
Luis Pater a325533f20 Merge pull request #2972 from XYenon/feat/amp-thread-id
feat: support X-Amp-Thread-Id for session affinity
2026-04-26 23:30:12 +08:00
sususu98 6fc23568df logging: mark antigravity credits requests 2026-04-26 23:04:27 +08:00
Luis Pater 736ae61e4a Merge pull request #3051 from philipbankier/fix/gpt55-free-tier-test
fix(test): remove free tier from GPT-5.5 inclusion test
2026-04-26 22:35:59 +08:00
Luis Pater c5bea6f6f8 Merge pull request #3020 from Matthias319/fix/codex-error-classification
fix(codex): classify context, thinking-signature, previous-response, and auth failures
2026-04-26 22:26:40 +08:00
Luis Pater c7b28ba058 feat(executor): add support for Codex image generation tool usage tracking
- Introduced `publishCodexImageToolUsage` to report image generation tool metrics.
- Updated executor logic to handle image generation tool events and defaults.
- Added parsing logic for `image_gen` tool usage details in `helps/usage_helpers.go`.
- Updated `UsageReporter` for additional model-specific usage publishing.
- Refactored usage detail normalizations.

Closes: #3063
2026-04-26 22:19:03 +08:00
Luis Pater 38573050aa feat(config): add support for disabling OpenAI compatibility providers
- Introduced a `Disabled` flag to OpenAI compatibility configurations.
- Updated routing, auth selection, and API handling logic to respect the `Disabled` state.
- Extended relevant APIs, YAML configurations, and data structures to include the `Disabled` field.
- Adjusted all relevant loops and filters to skip disabled providers.

Closes: #3060 #3059 #2977
2026-04-26 21:49:36 +08:00
RCDandCopilot 153d8c160f chore: merge upstream CLIProxyAPI v6.9.38
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-04-26 08:51:31 +00:00
philipbankier 32ef1588e8 fix(test): remove free tier from GPT-5.5 inclusion test
GPT-5.5 was correctly removed from codex-free tier in 7b89583c
(since free accounts cannot access it), but the test was not updated
to reflect this. This caused TestCodexStaticModelsIncludeGPT55 to
fail on the free subtest.

Changes:
- Remove free tier from GPT-5.5 inclusion test
- Add new TestCodexFreeModelsExcludeGPT55 to explicitly verify
  that free tier does NOT include GPT-5.5
2026-04-25 22:11:08 -04:00
edlsh 80eb03709a fix(openai): preserve multiline repaired SSE data 2026-04-25 18:12:27 -04:00
edlsh d36e70e9dc fix(openai): preserve unindexed response output items 2026-04-25 18:06:00 -04:00
edlsh fd45dece7f fix(openai): repair empty responses stream output 2026-04-25 17:46:44 -04:00
edlsh 95318ad46d fix(amp): preserve lowercase glob tool name 2026-04-25 17:46:44 -04:00
Enzo Lucchesi fc1ddf365f fix(claude): centralize oauth tool-name transform flow 2026-04-25 17:45:03 -04:00
edlsh 03ea4e569f perf(claude): pre-allocate reverseMap capacity
Address Gemini code review suggestion: the reverseMap can contain at
most len(oauthToolRenameMap) entries, so pre-allocating avoids
reallocations as entries are added.
2026-04-25 17:45:03 -04:00
Enzo Lucchesi e707cf7d46 fix(claude): only reverse-remap OAuth tool names that were forward-renamed
remapOAuthToolNames renames lowercase client-sent tools (e.g. `glob` →
`Glob`) to Claude Code equivalents on OAuth requests to avoid tool-name
fingerprinting. The reverse pass previously ran against a *global*
reverse map and rewrote every tool_use block whose name matched any
value in oauthToolRenameMap — regardless of what the client actually
sent.

For clients that send mixed casing (notably Amp CLI — `Bash`, `Read`,
`Grep`, `Task` alongside `glob`, `skill`, etc.) this corrupted the
response. Any forward rename in the request set the "renamed" flag,
which then unconditionally lowercased every `Bash` in the response to
`bash`. Amp's tool registry has `Bash`, not `bash`, so it rejected the
tool_use with `tool "bash" is not allowed for smart mode` and tool
execution failed.

Fix: `remapOAuthToolNames` now returns a per-request map keyed on the
upstream (TitleCase) name valued with the original client-sent name.
The reverse functions take this map and only touch entries in it.
Names the client sent in TitleCase pass through untouched in both
directions.

- Change remapOAuthToolNames signature from `([]byte, bool)` to
  `([]byte, map[string]string)`; populate at every rename site
  (tools[], tool_choice.name, message tool_use, tool_reference,
  nested tool_reference inside tool_result).
- Change reverseRemapOAuthToolNames and
  reverseRemapOAuthToolNamesFromStreamLine to accept and consume the
  per-request map; remove the global oauthToolRenameReverseMap.
- Update all three executor call sites (Execute, ExecuteStream direct
  passthrough, ExecuteStream translated) + count_tokens.
- Add regression tests for the mixed-case scenario in both the
  non-streaming and SSE code paths.
2026-04-25 17:45:03 -04:00
Luis Pater 0a7c6b0a4a feat(api): enhance model assignment logic in image handlers
- Updated `buildImagesResponsesRequest` to derive `model` dynamically based on `toolJSON`.
- Adjusted streaming execution to handle dynamic model resolution across multiple contexts.

Closes: #2965
2026-04-26 03:24:43 +08:00
Luis Pater ea670ef8c0 feat(models): add Codex Auto Review model entry to registry JSON
Closes: #2995
2026-04-26 03:09:06 +08:00
Luis Pater 2c626efc59 feat(security): implement IP ban for repeated management key and Redis AUTH failures
- Added IP ban logic to `AuthenticateManagementKey` and Redis protocol handlers, blocking requests after multiple failed attempts.
- Introduced unit tests to validate IP ban behavior across localhost and remote clients.
- Synchronized Redis protocol's authentication policy with management key validation.
2026-04-25 21:39:58 +08:00
Luis Pater 28d78273e4 feat(api): implement protocol multiplexer and Redis queue for usage integration
- Added `protocol_multiplexer.go`, enabling support for both HTTP and Redis protocols on a single listener.
- Introduced `redis_queue_protocol.go` to handle Redis-compatible RESP commands for queue management.
- Integrated `redisqueue` package, supporting in-memory queuing with expiration pruning.
- Updated server initialization to manage a shared listener and multiplex connections.
- Adjusted `Handler` to adopt `AuthenticateManagementKey` for modular key validation, supporting both HTTP and Redis flows.
2026-04-25 18:52:24 +08:00
Luis Pater be0fe6fab3 Merge pull request #3022 from Willxup/main
docs: add CPA Usage Keeper to ecosystem list
2026-04-25 11:44:49 +08:00
Will cf043f6c07 docs:Add CPA Usage Keeper to README ecosystem list 2026-04-24 23:54:09 +08:00
Will faad8e30dd Add CPA Usage Keeper to README ecosystem list 2026-04-24 23:28:44 +08:00
Luis Pater a7e92e2639 feat(auth): disallow free-tier Codex auth during selection process
- Introduced `disallowFreeAuthFromMetadata` and `isFreeCodexAuth` to enforce skipping free-tier credentials.
- Modified scheduler logic to honor `DisallowFreeAuthMetadataKey` during auth selection.
- Updated `ensureImageGenerationTool` to skip tool injection for free-tier Codex auth.
- Added context utility `WithDisallowFreeAuth` and integrated with image handlers.
- Augmented relevant tests to cover free-tier exclusion scenarios.
2026-04-24 23:18:56 +08:00
Matthias319 4056c2590b fix(codex): classify known upstream failures
Normalize Codex context, thinking-signature, previous-response, and auth failures to explicit error codes: context_too_large, thinking_signature_invalid, previous_response_not_found, auth_unavailable.

Refs #2596.
2026-04-24 17:13:23 +02:00
sususu98 36cc762fc9 Merge pull request #3007 from sususu98/fix-antigravity-credits-stream-fallback
fix(antigravity): trigger credits fallback for streaming
2026-04-24 17:28:47 +08:00
sususu98 5f5d5936fa fix antigravity credits stream fallback 2026-04-24 15:47:18 +08:00
RCD 35a58873b0 chore: merge upstream CLIProxyAPI v6.9.36 2026-04-24 04:25:06 +00:00
Luis Pater f1ba6151a9 feat(codex): pass base model to enable conditional image_generation tool injection
- Modified `ensureImageGenerationTool` to accept `baseModel` for conditional logic.
- Ensured `gpt-5.3-codex-spark` models bypass image_generation tool injection.
- Updated relevant tests and executor logic to reflect changes.
2026-04-24 07:21:03 +08:00
Luis Pater 7b89583cf8 chore(models): remove GPT-5.5 model entry from registry JSON 2026-04-24 05:07:03 +08:00
Luis Pater 1576d14137 Merge pull request #2989 from ben-vargas/gpt-5-5-support
Align GPT-5.5 Codex metadata with runtime cache
2026-04-24 04:24:58 +08:00
Ben Vargas 736018a0b0 Add GPT-5.5 Codex model support 2026-04-23 13:43:02 -06:00
Luis Pater 7d5f6d9382 feat(models): add GPT-5.5 model entry to registry JSON 2026-04-24 02:43:12 +08:00
sususu98 12195a276e Merge pull request #2971 from sususu98/feat/antigravity-credits-fallback
feat(antigravity): conductor-level credits fallback for Claude models
2026-04-24 00:15:23 +08:00
Luis Pater 25137b1984 feat(logging): add AI API path support for image routes
- Included `/v1/images` in AI API path prefixes.
- Introduced tests to validate `/v1/images/generations` and `/v1/images/edits` as AI API paths.
2026-04-24 00:11:42 +08:00
sususu98 7ad1900041 perf(antigravity): async credits hint refresh for warm tokens 2026-04-23 23:58:10 +08:00
Luis Pater 8eb56e5602 Merge pull request #2962 from MoYeRanqianzhi/feat/codex-image-generation-tool-injection
feat(codex): inject image_generation tool + route aliases for Codex CLI image generation
2026-04-23 23:48:31 +08:00
sususu98 f130846ec1 fix(auth): break credits cold-start deadlock by keeping unknown-hint auths as fallback candidates
Replace antigravityCreditsAvailableForModel with inline known/unknown
split. Auths whose credit hints are not yet populated are kept as
lower-priority candidates instead of being rejected, breaking the
chicken-and-egg deadlock at cold start.
2026-04-23 22:47:51 +08:00
sususu98 920b6efffa refactor(logging): strip unrelated deferred body changes, keep credits-only logging
Remove deferred body optimization and maxErrorLog constants that were
unrelated to credits fallback. Keep only MarkCreditsUsed/CreditsUsed
helpers for flagging requests that consumed AI credits.
2026-04-23 17:41:54 +08:00
sususu98 e75daa299b fix(antigravity): respect pinned auth in credits fallback, release deferred body on success
- findAllAntigravityCreditsCandidateAuths now filters by PinnedAuthMetadataKey
  to prevent credential isolation violations during credits fallback
- Release deferredBody reference on success path to avoid holding large
  payloads in memory for the lifetime of the gin context
2026-04-23 17:38:02 +08:00
XYenon 8e49c795f5 fix: forward HTTP headers to executor Options so session affinity can read X-Amp-Thread-Id 2026-04-23 15:34:31 +08:00
sususu98 4de5c29f86 fix(antigravity): remove credits fallback from CountTokens, fix gofmt
CountTokens upstream API does not support enabledCreditTypes, so
remove the dead credits fallback path from ExecuteCount and delete
the unused tryAntigravityCreditsExecuteCount method. Fix gofmt on
credits test file.
2026-04-23 15:17:00 +08:00
XYenon 4d6457e6ec feat: support extracting X-Amp-Thread-Id header as session id for session affinity 2026-04-23 13:49:00 +08:00
sususu98 14d46a0a5d feat(antigravity): conductor-level credits fallback for Claude models
Move credits handling from executor-level retry to conductor-level
orchestration. When all free-tier auths are exhausted (429/503), the
conductor discovers auths with available Google One AI credits and
retries with enabledCreditTypes injected via context flag.

Key changes:
- Add AntigravityCreditsHint system for tracking per-auth credits state
- Conductor tries credits fallback after all auths fail (Execute/Stream/Count)
- Executor injects enabledCreditTypes only when conductor sets context flag
- Credits fallback respects provider scope (requires antigravity in providers)
- Add context cancellation check in credits fallback to avoid wasted requests
- Remove executor-level attemptCreditsFallback and preferCredits machinery
- Restructure 429 decision logic (parse details first, keyword fallback)
- Expand shouldAbort to cover INVALID_ARGUMENT/FAILED_PRECONDITION/500+UNKNOWN
- Support human-readable retry delay parsing (e.g. "1h43m56s")
2026-04-23 13:44:20 +08:00
MoYeRanQianZhi 31934ae04c feat(codex): enable image generation for all Codex upstream requests
Codex CLI gates the built-in image_generation tool behind
AuthMode::Chatgpt (OAuth only). When clients connect via API key
auth through CPA, the tool is absent from requests, making image
generation unavailable through the reverse proxy.

Changes:

1. Inject image_generation tool (codex_executor.go):
   Add ensureImageGenerationTool() that appends
   {"type":"image_generation","output_format":"png"} to the tools
   array if not already present. Applied to all three execution
   paths: Execute, executeCompact, and ExecuteStream.

2. Route aliases for Codex CLI direct access (server.go):
   Add /backend-api/codex/responses routes that map to the same
   OpenAI Responses API handlers as /v1/responses. This allows
   Codex CLI to connect via chatgpt_base_url config while keeping
   AuthMode::Chatgpt, which enables the built-in image_generation
   tool on the client side.

3. Unit tests (codex_executor_imagegen_test.go):
   Cover no-tools, existing tools, already-present, empty array,
   and mixed built-in tool scenarios.
2026-04-23 01:24:40 +08:00
Luis Pater a188159632 fix(handlers): remove references to unsupported n parameter in OpenAI image handlers 2026-04-22 21:28:17 +08:00
Luis Pater fd71960c3e fix(handlers): remove handling of unsupported n parameter in OpenAI image handlers 2026-04-22 21:12:50 +08:00
Luis Pater e935196df4 feat(models): add hardcoded GPT-Image-2 model support in Codex
- Added `GPT-Image-2` as a built-in model to avoid dependency on remote updates for Codex.
- Updated model tier functions (`CodexFree`, `CodexTeam`, etc.) to include built-in models via `WithCodexBuiltins`.
- Introduced new handlers for image generation and edit operations under `OpenAIAPIHandler`.
- Extended tests to validate 503 response for unsupported image model requests.
2026-04-22 20:51:13 +08:00
Luis Pater 4fc2c619fb feat(models): add Kimi K2.6 model entry to registry JSON 2026-04-21 20:53:03 +08:00
Luis Pater 8ced7a548f Merge pull request #2834 from muzhi1991/fix/openai-compat-host-header
fix(util): forward custom Host header to upstream
2026-04-21 20:29:14 +08:00
Luis Pater 3444820958 Merge pull request #2939 from stringer07/fix/codex-stream-output-backfill
fix(codex): backfill streaming response output
2026-04-21 20:18:57 +08:00
Luis Pater 1716a845eb feat(api): add support for HEAD requests to /healthz endpoint
- Refactored `/healthz` handler to support `HEAD` requests alongside `GET`.
- Updated tests to include validation for `HEAD` requests with expected status and empty body.

Closes: #2929
2026-04-21 20:16:18 +08:00
stringer07 b6781d69be perf(codex): avoid repeated output patch writes 2026-04-21 16:29:54 +08:00
stringer07 bb8408cef5 fix(codex): backfill streaming response output 2026-04-21 16:03:56 +08:00
Luis Pater e6866ff19c feat(auth): add refresh backoff for ineffective token updates
- Introduced `refreshIneffectiveBackoff` to prevent tight-looping in auto-refresh when token refresh fails to update expiry.
- Adjusted refresh logic to apply backoff when `shouldRefresh` evaluates true.

Closes: #2830
2026-04-20 15:40:43 +08:00
RCD 2ebbb59670 chore: merge upstream CLIProxyAPI v6.9.30
# Conflicts:
#	README.md
#	README_CN.md
#	README_JA.md
2026-04-20 12:48:28 +07:00
Luis Pater 8f4a4eabfc feat(docs): add VisionCoder sponsorship details and optimize external links
- Added VisionCoder sponsorship information to `README.md`, `README_CN.md`, and `README_JA.md`.
- Updated external links to include `target="_blank"` for improved user experience.
- Added new logo asset `visioncoder.png` for README use.
2026-04-19 23:00:09 +08:00
Luis Pater e05abec01f Merge pull request #2898 from octo-patch/fix/issue-2866-remove-obsolete-1m-beta-header
fix(executor): drop obsolete context-1m-2025-08-07 beta header
2026-04-19 11:46:21 +08:00
octo-patch f4eb16102b fix(executor): drop obsolete context-1m-2025-08-07 beta header (fixes #2866)
Anthropic has moved the 1M-context-window feature to General Availability,
so the context-1m-2025-08-07 beta flag is no longer accepted and now causes
400 Bad Request errors when forwarded upstream.

Remove the X-CPA-CLAUDE-1M detection and the corresponding injection of the
now-invalid beta header.  Also drop the unused net/textproto import that was
only needed for the header-key lookup.
2026-04-19 10:38:16 +08:00
Luis Pater 86c856f56f feat(translator): add partial and full image generation support in Codex-GPT and Codex-Gemini flows
- Introduced `LastImageHashByItemID` in Codex-GPT and `LastImageHashByID` in Codex-Gemini for deduplication of generated images.
- Added support for handling `partial_image` and `image_generation_call` types, with inline data embedding for Gemini and URL payload conversion for GPT.
- Extended unit tests to verify image handling in both streaming and non-streaming modes.
2026-04-19 03:21:59 +08:00
RCD 2221b8e557 chore: merge upstream CLIProxyAPI v6.9.29 2026-04-18 23:28:27 +07:00
Supra4E8C c6baa64b4e Merge pull request #2892 from router-for-me/fix-provider
feat(api): integrate auth index into key retrieval endpoints for Gemi…
2026-04-18 17:24:43 +08:00
Supra4E8C a64141a9a6 fix(tests): remove obsolete config_auth_index_test file 2026-04-18 17:22:16 +08:00
Supra4E8C c26936e2e6 fix(management): stabilize auth-index mapping 2026-04-18 17:12:14 +08:00
Supra4E8C 894baad829 feat(api): integrate auth index into key retrieval endpoints for Gemini, Claude, Codex, OpenAI, and Vertex 2026-04-18 16:44:33 +08:00
muzhi1991 eba561bf6f fix(util): also keep Host in header map for synthetic requests
Addressing the P1 note from the Codex reviewer: applyCustomHeaders is
also called with a synthetic &http.Request{Header: ...} from the
websockets executors (aistudio_executor.go, codex_websockets_executor.go),
which forward only the header map. The previous continue meant a custom
Host was dropped from that map, regressing virtual-host overrides on
those flows. Mirror the value to both r.Host (for real net/http) and
r.Header (for header-map-only consumers).
2026-04-17 09:28:59 +08:00
hkfires da43f63735 fix(tests): update Gemini family test case numbers for consistency 2026-04-17 08:43:19 +08:00
hkfires d9a3b3e5f3 fix(tests): update model lookup references and enhance Claude executor tests 2026-04-17 08:32:07 +08:00
RCD 7ec328f952 chore: merge upstream CLIProxyAPI v6.9.28 2026-04-17 07:04:22 +07:00
Luis Pater 5dcca69e8c feat(models): add Claude Opus 4.7 model entry to registry JSON 2026-04-17 01:08:19 +08:00
Luis Pater f5dc6483d5 chore: remove iFlow-related modules and dependencies
- Deleted `iflow` provider implementation, including thinking configuration (`apply.go`) and authentication modules.
- Removed iFlow-specific tests, executors, and helpers across SDK and internal components.
- Updated all references to exclude iFlow functionality.
2026-04-17 01:07:12 +08:00
Luis Pater d949921143 feat(auth): add proxy URL override support to auth constructors and executors
- Introduced `WithProxyURL` variants for `CodexAuth`, `ClaudeAuth`, `IFlowAuth`, and `DeviceFlowClient`.
- Updated executors to use proxy-aware constructors for improved configurability.
- Added unit tests to validate proxy override precedence and functionality.

Closes: #2823
2026-04-16 22:11:39 +08:00
Luis Pater 7b03f04670 fix(handlers): include execution session metadata and skip idempotency key when absent
- Refactored `requestExecutionMetadata` to handle empty `Idempotency-Key` gracefully.
- Added test to validate metadata inclusion of execution session without idempotency key.
2026-04-16 21:44:32 +08:00
muzhi1991 8f9e6622b0 fix(util): forward custom Host header to upstream
Custom headers configured under openai-compatibility (and any other
provider passing through applyCustomHeaders) were silently dropped for
the Host key, because Go's net/http reads the wire Host from
req.Host, not req.Header["Host"]. As a result, virtual-host routed
upstreams (e.g. LiteLLM behind an ingress) saw the base-url's host
instead of the user-configured override and returned 404.

Detect the Host key with http.CanonicalHeaderKey and assign it to
req.Host so it is actually written on the wire. Other headers continue
to use Header.Set as before.

Fixes #2833
2026-04-16 20:45:37 +08:00
RCD 71be5e897b chore: merge upstream CLIProxyAPI v6.9.27 2026-04-16 09:00:13 +07:00
hkfires 1267fddf61 fix(docker-build): improve argument handling and error messaging for usage option 2026-04-16 09:19:03 +08:00
Luis Pater ba454dbfbf Merge pull request #2817 from sususu98/fix/antigravity-strip-billing-header
fix(antigravity): strip billing header from system instruction before upstream call
2026-04-16 02:46:05 +08:00
Luis Pater d1508ca030 Merge pull request #2816 from sususu98/feat/session-affinity
feat(session-affinity): add session-sticky routing for multi-account load balancing
2026-04-16 02:45:31 +08:00
sususu98 d4a6a5ae15 fix(antigravity): strip billing header from system instruction before upstream call
The x-anthropic-billing-header block in the Claude system array is
client-internal metadata and should not be forwarded to the Gemini
upstream as part of systemInstruction.parts.
2026-04-16 00:19:01 +08:00
sususu98 7c24d54ca8 feat(session-affinity): add session-sticky routing for multi-account load balancing
When multiple auth credentials are configured, requests from the same
session are now routed to the same credential, improving upstream prompt
cache hit rates and maintaining context continuity.

Core components:
- SessionAffinitySelector: wraps RoundRobin/FillFirst selectors with
  session-to-auth binding; automatic failover when bound auth is
  unavailable, re-binding via the fallback selector for even distribution
- SessionCache: TTL-based in-memory cache with background cleanup
  goroutine, supporting per-session and per-auth invalidation
- StoppableSelector interface: lifecycle hook for selectors holding
  resources, called during Manager.StopAutoRefresh()

Session ID extraction priority (extractSessionIDs):
1. metadata.user_id with Claude Code session format (old
   user_{hash}_session_{uuid} and new JSON {session_id} format)
2. X-Session-ID header (generic client support)
3. metadata.user_id (non-Claude format, used as-is)
4. conversation_id field
5. Stable FNV hash from system prompt + first user/assistant messages
   (fallback for clients with no explicit session ID); returns both a
   full hash (primaryID) and a short hash without assistant content
   (fallbackID) to inherit bindings from the first turn

Multi-format message hash covers OpenAI messages, Claude system array,
Gemini contents/systemInstruction, and OpenAI Responses API input items
(including inline messages with role but no type field).

Configuration (config.yaml routing section):
- session-affinity: bool (default false)
- session-affinity-ttl: duration string (default "1h")
- claude-code-session-affinity: bool (deprecated, alias for above)
All three fields trigger selector rebuild on config hot reload.

Side effect: Idempotency-Key header is no longer auto-generated with a
random UUID when absent — only forwarded when explicitly provided by the
client, to avoid polluting session hash extraction.
2026-04-16 00:18:47 +08:00
Luis Pater a4c1e32ff6 chore(models): remove outdated GPT-5 and related model entries from registry JSON 2026-04-15 20:37:32 +08:00
Luis Pater f56cf42461 Merge pull request #2800 from sususu98/fix/antigravity-max-output-tokens-cap
fix(antigravity): cap maxOutputTokens using registry max_completion_tokens
2026-04-15 20:35:11 +08:00
Luis Pater 3dea1da249 Merge pull request #2782 from sususu98/fix/strip-invalid-signature-thinking-blocks
fix(antigravity): use E-prefixed fake signature in strict bypass test
2026-04-15 20:34:32 +08:00
Luis Pater 8fac29631d chore: remove Qwen support from SDK and internal components
- Deleted `QwenAuthenticator`, internal `qwen_auth`, and `qwen_executor` implementations.
- Removed all Qwen-related OAuth flows, token handling, and execution logic.
- Cleaned up dependencies and references to Qwen across the codebase.
2026-04-15 12:16:08 +08:00
sususu98 8fecd625d2 fix(antigravity): cap maxOutputTokens using registry max_completion_tokens
Claude models on antigravity have a 64000 token output limit but
max_tokens from downstream requests was passed through uncapped,
causing 400 INVALID_ARGUMENT from Google when clients sent 128000.
2026-04-15 11:57:55 +08:00
sususu98 10b55b5ddd fix(antigravity): use E-prefixed fake signature in strict bypass test
The strict bypass test used testGeminiSignaturePayload() which produces
a base64 string starting with 'C'. Since StripInvalidSignatureThinkingBlocks
now strips all non-E/R signatures unconditionally, the test payload was
stripped before reaching ValidateClaudeBypassSignatures, causing the test
to pass the request through instead of rejecting it with 400.

Replace with testFakeClaudeSignature() which produces a base64 string
starting with 'E' (valid at the lightweight check) but with invalid
protobuf content (no valid field 2), so strict mode correctly rejects
it at the deep validation layer.
2026-04-14 15:46:02 +08:00
sususu98 41ae2c81e7 fix(antigravity): discard thinking blocks with non-Claude-format signatures
Proxy-generated thinking blocks may carry hex hashes or other non-Claude
signatures (e.g. "d5cb9cd0823142109f451861") from Gemini responses. These
are now discarded alongside empty-signature blocks during the strip phase,
before validation runs. Valid Claude signatures always start with 'E' or 'R'
(after stripping any cache prefix).
2026-04-14 15:14:48 +08:00
sususu98 278a89824c fix(antigravity): strip thinking blocks with empty signatures instead of rejecting
Thinking blocks with empty signatures come from proxy-generated
responses (Antigravity/Gemini routed as Claude). These should be
silently dropped from the request payload before forwarding, not
rejected with 400. Fixes 10 "missing thinking signature" errors.
2026-04-14 15:14:48 +08:00
RCD 06c3e95454 chore: update upstream markers for v6.9.24 2026-04-13 05:29:08 +07:00
RCD a6cc1a2e71 Merge tag 'v6.9.24' into sync-upstream-v6.9.23-to-v6.9.24 2026-04-13 05:28:14 +07:00
sususu98 c4459c4346 Merge pull request #2724 from sususu98/fix/skip-schema-cleanup-empty-tools
fix(antigravity): skip full schema cleanup for empty tool requests
2026-04-12 14:05:47 +08:00
sususu98 61e0447f92 Merge pull request #2723 from sususu98/fix/drop-redacted-thinking-blocks
fix(antigravity): drop redacted thinking blocks with empty text
2026-04-12 14:05:41 +08:00
sususu98 1dc3018fd6 Merge pull request #2716 from sususu98/pr/antigravity-bypass-log-noise
fix(antigravity): reduce bypass mode log noise
2026-04-12 14:05:34 +08:00
sususu98 26fd3eff03 Merge pull request #2715 from sususu98/pr/antigravity-32mb-bypass-signatures
fix(antigravity): allow 32MB bypass signatures
2026-04-12 14:05:27 +08:00
Luis Pater 5bfaf8086b feat(auth): add configurable worker pool size for auto-refresh loop
- Introduced `auth-auto-refresh-workers` config option to override default concurrency.
- Updated `authAutoRefreshLoop` to support customizable worker counts.
- Enhanced token refresh scheduling flexibility by aligning worker pool with runtime configurations.
2026-04-12 13:56:05 +08:00
Luis Pater 6c0a1efd71 refactor(auth): simplify auth directory scanning and improve JSON processing logic
- Replaced `filepath.Walk` with `os.ReadDir` for cleaner directory traversal.
- Fixed `isAuthJSON` check to use `filepath.Dir` for directory comparison.
- Updated auth hash cache generation and file synthesis to improve readability and maintainability.
2026-04-12 13:32:03 +08:00
sususu98andSisyphus f5ed5c7453 fix(antigravity): skip full schema cleanup for empty tool requests
Avoid whole-payload schema sanitization when translated Antigravity requests have no actual tool schemas, including missing and empty tools arrays. Add regression coverage so image-heavy no-tool requests keep bypassing the old memory amplification path.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-12 12:51:42 +08:00
sususu98 65158cce46 fix(antigravity): drop redacted thinking blocks with empty text
Antigravity wraps empty thinking text into a prompt-caching-scope
object that omits the required inner "thinking" field, causing 400
"messages.N.content.0.thinking.thinking: Field required" when Claude
Max requests are routed through Antigravity in bypass mode.
2026-04-12 12:30:43 +08:00
RCD 9cc006c941 chore: merge upstream CLIProxyAPI v6.9.23 2026-04-12 02:15:21 +07:00
Luis Pater a583463d60 feat(auth): implement auto-refresh loop for managing auth token schedule
- Introduced `authAutoRefreshLoop` to handle token refresh scheduling.
- Replaced semaphore-based refresh logic in `Manager` with the new loop.
- Added unit tests to verify refresh schedule logic and edge cases.
2026-04-12 02:06:40 +08:00
sususu98andSisyphus 8ed290c1c4 fix(antigravity): reduce bypass mode log noise
Keep cache-disable visibility at info level while suppressing duplicate state-change logs and moving strict-mode chatter down to debug.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-12 00:52:41 +08:00
sususu98andSisyphus 727221df2e fix(antigravity): allow 32MB bypass signatures
Raise the local bypass-signature ceiling so long Claude thinking signatures are not rejected before request translation, and keep the oversized-signature test cheap to execute.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-12 00:51:53 +08:00
Luis Pater 0ab1f5412f fix(executor): handle 429 Retry-After header and default retry logic for quota exhaustion
- Added proper parsing of `Retry-After` headers for 429 responses.
- Set default retry duration when "disable cooling" is active on quota exhaustion.
- Updated tests to verify `Retry-After` handling and default behavior.
2026-04-11 21:04:55 +08:00
Luis Pater 9ded75d335 Merge pull request #2702 from AllenReder/docs/add-quota-inspector
docs(README): add CLIproxyAPI Quota Inspector to community projects list
2026-04-11 16:42:02 +08:00
Allen Yi f135fdf7fc docs: clarify codex quota window wording in README locales 2026-04-11 16:39:32 +08:00
Luis Pater 828df80088 refactor(executor): remove immediate retry with token refresh on 429 for Qwen and update tests accordingly 2026-04-11 16:35:18 +08:00
Allen Yi c585caa0ce docs: fix CLIProxyAPI Quota Inspector naming and link casing 2026-04-11 16:22:45 +08:00
Allen Yi 5bb69fa4ab docs: refine CLIproxyAPI Quota Inspector description in all README locales 2026-04-11 15:22:27 +08:00
RCD 85267e7379 fix: resolve v6.9.22 cache metric cycles 2026-04-11 00:10:43 +07:00
RCD 45dc09d487 chore: merge upstream CLIProxyAPI v6.9.22 2026-04-11 00:06:28 +07:00
Luis Pater 5ab9afac83 fix(executor): handle OAuth tool name remapping with rename detection and add tests
Closes: #2656
2026-04-10 21:54:59 +08:00
Luis Pater 65ce86338b fix(executor): implement immediate retry with token refresh on 429 for Qwen and add associated tests
Closes: #2661
2026-04-10 21:12:03 +08:00
Chén Mù 2a97037d7b Merge pull request #2670 from sususu98/feat/antigravity-prefer-prod-url
feat(antigravity): prefer prod URL as first priority
2026-04-10 19:43:27 +08:00
sususu98 d801393841 feat(antigravity): prefer prod URL as first priority
Promote cloudcode-pa.googleapis.com to the first position in the
fallback order, with daily and sandbox URLs as fallbacks.
2026-04-10 19:37:56 +08:00
Luis Pater b2c0cdfc88 Merge pull request #2621 from wykk-12138/fix/oauth-extra-usage-detection
fix(claude): prevent OAuth extra-usage billing via tool name fingerprinting and system prompt cloaking
2026-04-10 10:29:27 +08:00
Luis Pater f32c8c9620 fix(handlers): update listener to bind on all interfaces instead of localhost
Fixed: #2640
2026-04-10 07:24:34 +08:00
wykk-12138andSisyphus 0f45d89255 fix(claude): address PR review feedback for OAuth cloaking
- Use buildTextBlock for billing header to avoid raw JSON string interpolation
- Fix empty array edge case in prependToFirstUserMessage
- Allow remapOAuthToolNames to process messages even without tools array
- Move claude_system_prompt.go to helps/ per repo convention
- Export prompt constants (ClaudeCode* prefix) for cross-package access

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-10 00:07:11 +08:00
wykk-12138 96056d0137 Merge remote-tracking branch 'upstream/main' into fix/oauth-extra-usage-detection 2026-04-09 22:59:31 +08:00
wykk-12138andSisyphus f780c289e8 fix(claude): map question/skill to TitleCase instead of removing them
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-09 22:28:00 +08:00
wykk-12138andSisyphus ac36119a02 fix(claude): preserve OAuth tool renames when filtering tools
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-09 22:20:15 +08:00
Luis Pater 39dc4557c1 Merge pull request #2412 from sususu98/feat/signature-cache-toggle
feat: configurable signature cache toggle for Antigravity/Claude thinking blocks
2026-04-09 21:54:47 +08:00
ZTXBOSS666 30e94b6792 fix(antigravity): refine 429 handling and credits fallback
Includes: restore SDK docs under docs/; update antigravity executor credits tests; gofmt.
2026-04-09 21:48:32 +08:00
sususu98 38f0ae5970 docs(antigravity): document signature validation spec alignment
Add package-level comment documenting the protobuf tree structure,
base64 encoding equivalence proof, output dimensions, and spec
section references. Remove unreachable legacy_vertex_group dead code.
2026-04-09 21:12:40 +08:00
sususu98 cf249586a9 feat(antigravity): configurable signature cache with bypass-mode validation
Antigravity 的 Claude thinking signature 处理新增 cache/bypass 双模式,
并为 bypass 模式实现按 SIGNATURE-CHANNEL-SPEC.md 的签名校验。

新增 antigravity-signature-cache-enabled 配置项(默认 true):
- cache mode(true):使用服务端缓存的签名,行为与原有逻辑完全一致
- bypass mode(false):直接使用客户端提供的签名,经过校验和归一化

支持配置热重载,运行时可切换模式。

校验流程:
1. 剥离历史 cache-mode 的 'modelGroup#' 前缀(如 claude#Exxxx → Exxxx)
2. 首字符必须为 'E'(单层编码)或 'R'(双层编码),否则拒绝
3. R 开头:base64 解码 → 内层必须以 'E' 开头 → 继续单层校验
4. E 开头:base64 解码 → 首字节必须为 0x12(Claude protobuf 标识)
5. 所有合法签名归一化为 R 形式(双层 base64)发往 Antigravity 后端

非法签名处理策略:
- 非严格模式(默认):translator 静默丢弃无签名的 thinking block
- 严格模式(antigravity-signature-bypass-strict: true):
  executor 层在请求发往上游前直接返回 HTTP 400

按 SIGNATURE-CHANNEL-SPEC.md 解析 Claude 签名的完整 protobuf 结构:
- Top-level Field 2(容器)→ Field 1(渠道块)
- 渠道块提取:channel_id (Field 1)、infrastructure (Field 2)、
  model_text (Field 6)、field7 (Field 7)
- 计算 routing_class、infrastructure_class、schema_features
- 使用 google.golang.org/protobuf/encoding/protowire 解析

- resolveThinkingSignature 拆分为 resolveCacheModeSignature / resolveBypassModeSignature
- hasResolvedThinkingSignature:mode-aware 签名有效性判断
  (cache: len>=50 via HasValidSignature,bypass: non-empty)
- validateAntigravityRequestSignatures:executor 预检,
  仅在 bypass + strict 模式下拦截非法签名返回 400
- 响应侧签名缓存逻辑与 cache mode 集成
- Cache mode 行为完全保留:无 '#' 前缀的原生签名静默丢弃
2026-04-09 21:12:40 +08:00
Luis Pater 1dba2d0f81 fix(handlers): add base URL validation and improve API key deletion tests 2026-04-09 20:51:54 +08:00
Luis Pater 730809d8ea fix(auth): preserve and restore ready view cursors during index rebuilds 2026-04-09 20:26:16 +08:00
wykk-12138 e8d1b79cb3 fix(claude): remap OAuth tool names to Claude Code style to avoid third-party fingerprint detection
A/B testing confirmed that Anthropic uses tool name fingerprinting to detect
third-party clients on OAuth traffic. OpenCode-style lowercase names like
'bash', 'read', 'todowrite' trigger extra-usage billing, while Claude Code
TitleCase names like 'Bash', 'Read', 'TodoWrite' pass through normally.

Changes:
- Add oauthToolRenameMap: maps lowercase tool names to Claude Code equivalents
- Add oauthToolsToRemove: removes 'question' and 'skill' (no Claude Code counterpart)
- remapOAuthToolNames: renames tools, removes blacklisted ones, updates tool_choice and messages
- reverseRemapOAuthToolNames/reverseRemapOAuthToolNamesFromStreamLine: reverse map for responses
- Apply in Execute(), ExecuteStream(), and CountTokens() for OAuth token requests
2026-04-09 20:15:16 +08:00
Luis Pater 5e81b65f2f fix(auth, executor): normalize Qwen base URL, adjust RefreshLead duration, and add tests 2026-04-09 18:07:07 +08:00
github-actions[bot] 5775476bc0 chore: sync CLIProxyAPI v6.9.15-v6.9.18 2026-04-09 09:15:47 +00:00
wykk-12138andSisyphus 7e8e2226a6 fix(claude): reduce forwarded OAuth prompt to minimal tool reminder
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-09 17:12:07 +08:00
wykk-12138andSisyphus f0c20e852f fix(claude): remove invalid cache_control scope from static system block
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-09 17:00:04 +08:00
wykk-12138 7cdf8e9872 fix(claude): sanitize forwarded third-party prompts for OAuth cloaking
Only for Claude OAuth requests, sanitize forwarded system-prompt context before
it is prepended into the first user message. This preserves neutral task/tool
instructions while removing OpenCode branding, docs links, environment banners,
and product-specific workflow sections that still triggered Anthropic extra-usage
classification after top-level system[] cloaking.
2026-04-09 16:45:29 +08:00
wykk-12138 e2e3c7dde0 fix: remove invalid org scope and match Claude Code block layout 2026-04-09 14:09:52 +08:00
wykk-12138 9e0ab4d116 fix: build cache_control JSON manually to avoid sjson map marshaling 2026-04-09 14:03:23 +08:00
wykk-12138 8783caf313 fix: buildTextBlock cache_control sjson path issue
sjson treats 'cache_control.type' as nested path, creating
{ephemeral: {scope: org}} instead of {type: ephemeral, scope: org}.
Pass the whole map to sjson.SetBytes as a single value.
2026-04-09 13:58:04 +08:00
wykk-12138 f6f4640c5e fix: use sjson to build system blocks, avoid raw newlines in JSON
The previous commit used fmt.Sprintf with %s to insert multi-line string
constants into JSON strings. Go raw string literals contain actual newline
bytes, which produce invalid JSON (control characters in string values).

Replace with buildTextBlock() helper that uses sjson.SetBytes to properly
escape text content for JSON serialization.
2026-04-09 13:50:49 +08:00
wykk-12138 613fe6768d fix(executor): inject full Claude Code system prompt blocks with proper cache scopes
Previous fix only injected billing header + agent identifier (2 blocks).
Anthropic's updated detection now validates system prompt content depth:
- Block count (needs 4-6 blocks, not 2)
- Cache control scopes (org for agent, global for core prompt)
- Presence of known Claude Code instruction sections

Changes:
- Add claude_system_prompt.go with extracted Claude Code v2.1.63 system prompt
  sections (intro, system instructions, doing tasks, tone & style, output efficiency)
- Rewrite checkSystemInstructionsWithSigningMode to build 5 system blocks:
  [0] billing header (no cache_control)
  [1] agent identifier (cache_control: ephemeral, scope=org)
  [2] core intro prompt (cache_control: ephemeral, scope=global)
  [3] system instructions (no cache_control)
  [4] doing tasks (no cache_control)
- Third-party client system instructions still moved to first user message

Follow-up to 69b950db4c
2026-04-09 12:58:50 +08:00
Luis Pater ad8e3964ff fix(auth): add retry logic for 429 status with Retry-After and improve testing 2026-04-09 07:07:19 +08:00
Luis Pater 941334da79 fix(auth): handle OAuth model alias in retry logic and refine Qwen quota handling 2026-04-09 03:44:19 +08:00
Luis Pater d54f816363 fix(executor): update Qwen user agent and enhance header configuration 2026-04-09 01:45:52 +08:00
wykk-12138 69b950db4c fix(executor): fix OAuth extra usage detection by Anthropic API
Three changes to avoid Anthropic's content-based system prompt validation:

1. Fix identity prefix: Use 'You are Claude Code, Anthropic's official CLI
   for Claude.' instead of the SDK agent prefix, matching real Claude Code.

2. Move user system instructions to user message: Only keep billing header +
   identity prefix in system[] array. User system instructions are prepended
   to the first user message as <system-reminder> blocks.

3. Enable cch signing for OAuth tokens by default: The xxHash64 cch integrity
   check was previously gated behind experimentalCCHSigning config flag.
   Now automatically enabled when using OAuth tokens.

Related: router-for-me/CLIProxyAPI#2599
2026-04-09 00:06:38 +08:00
hkfires 343a2fc2f7 docs: update AGENTS.md for improved clarity and detail in commands and architecture 2026-04-08 12:33:16 +08:00
Luis Pater 12b967118b Merge pull request #2592 from router-for-me/tests
fix(tests): update test cases
2026-04-08 11:57:15 +08:00
Luis Pater 70efd4e016 chore: add workflow to retarget main PRs to dev automatically 2026-04-08 10:35:49 +08:00
Luis Pater f5aa68ecda chore: add workflow to prevent AGENTS.md modifications in pull requests 2026-04-08 10:12:51 +08:00
hkfires d390b95b76 fix(tests): update test cases 2026-04-08 08:53:50 +08:00
Luis Pater d1f6224b70 Merge pull request #2569 from LucasInsight/fix/record-zero-usage
fix: record zero usage
2026-04-08 08:13:11 +08:00
Luis Pater fcc59d606d fix(translator): add unit tests to validate output_item.done fallback logic for Gemini and Claude 2026-04-08 03:54:15 +08:00
Luis Pater 91e7591955 fix(executor): add transient 429 resource exhausted handling with retry logic 2026-04-08 02:48:53 +08:00
Luis Pater c8b7e2b8d6 fix(executor): ensure empty stream completions use output_item.done as fallback
Fixed: #2583
2026-04-07 18:21:12 +08:00
DragonFSKY 4ca00f7983 fix(websocket): gate compact replay by downstream support 2026-04-07 14:25:05 +08:00
DragonFSKY d2d0e6f6a1 fix(websocket): narrow compact replay detection 2026-04-07 14:23:44 +08:00
DragonFSKY a0fe273081 fix(websocket): skip stale state merge after client-side compact
After a Codex CLI compact, the client sends a full conversation
transcript (with compaction items or assistant messages) as input.
Previously, normalizeResponseSubsequentRequest() unconditionally
merged this with stale lastRequest/lastResponseOutput, breaking
function_call/function_call_output pairings and causing 400 errors
("No tool output found for function call").

Add inputContainsFullTranscript() heuristic that detects compaction
items (type=compaction/compaction_summary) or assistant messages in
the input array, and bypasses the merge when a full transcript is
present.

Fixes #2207
2026-04-07 14:22:53 +08:00
Luis Pater cad45ffa33 Merge pull request #2578 from LemonZuo/feat_socks5h
feat: support socks5h scheme for proxy settings
2026-04-07 09:57:18 +08:00
Luis Pater 6a27bceec0 Merge pull request #2576 from zilianpn/fix/disable-cooling-auth-errors
fix(auth): honor disable-cooling and enrich no-auth errors
2026-04-07 09:56:25 +08:00
Lemon 163d68318f feat: support socks5h scheme for proxy settings 2026-04-07 07:46:11 +08:00
zilianpn 0ea768011b fix(auth): honor disable-cooling and enrich no-auth errors 2026-04-07 01:12:13 +08:00
Michael 8b9dbe10f0 fix: record zero usage 2026-04-06 20:19:42 +08:00
Code_GandSisyphus 29e32aaab9 fix(executor): route Claude refresh through retry-aware auth
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-06 17:16:42 +09:00
Code_GandSisyphus 6431cec7d3 fix(claude-auth): dedupe OAuth refresh and honor 429 backoff
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-06 17:16:15 +09:00
Luis Pater 9f5bdfaa31 Merge pull request #2531 from jamestut/openai-vertex-token-usage-fix
Fix missing `response.completed.usage` for late-usage OpenAI-compatible streams
2026-04-06 09:30:49 +08:00
Luis Pater 9eabdd09db Merge pull request #2522 from aikins01/fix/strip-tool-use-signature
fix(amp): strip signature from tool_use blocks before forwarding to Claude
2026-04-06 09:30:14 +08:00
Luis Pater c3f8dc362e Merge pull request #2491 from mpfo0106/feature/claude-code-safe-alignment-sentinels
test(claude): add compatibility sentinels and centralize builtin fallback handling
2026-04-06 09:27:08 +08:00
Luis Pater b85120873b Merge pull request #2332 from RaviTharuma/fix/claude-thinking-signature
fix: preserve Claude thinking signatures in Codex translator
2026-04-06 09:25:06 +08:00
Luis Pater 6f58518c69 docs(readme): remove redundant GITSTORE_GIT_BRANCH description in README files 2026-04-06 09:23:04 +08:00
Luis Pater 000fcb15fa Merge pull request #2298 from snoyiatk/feat/add-gitstore-branch
feat(gitstore): add support for specifying git branch (via GITSTORE_G…
2026-04-06 09:21:03 +08:00
Luis Pater ea43361492 Merge pull request #2121 from destinoantagonista-wq/main
Reconcile registry model states on auth changes
2026-04-06 09:13:27 +08:00
Luis Pater c1818f197b Merge pull request #1940 from Blue-B/fix/claude-interleaved-thinking-amp-gzip-budget
fix(claude): enable interleaved-thinking beta, decode AMP error gzip, fix budget 400
2026-04-06 09:08:23 +08:00
Aikins Laryea b0653cec7b fix(amp): strip signature from tool_use blocks before forwarding to Claude
ensureAmpSignature injects signature:"" into tool_use blocks so the
Amp TUI does not crash on P.signature.length. when Amp sends the
conversation back, Claude rejects the extra field with 400:
  tool_use.signature: Extra inputs are not permitted

strip the proxy-injected signature from tool_use blocks in
SanitizeAmpRequestBody before forwarding to the upstream API.
2026-04-05 12:26:24 +00:00
Luis Pater 22a1a24cf5 feat(executor): add tests for preserving key order in cache control functions
Added comprehensive tests to ensure key order is maintained when modifying payloads in `normalizeCacheControlTTL` and `enforceCacheControlLimit` functions. Removed unused helper functions and refactored implementations for better readability and efficiency.
2026-04-05 17:58:13 +08:00
RCD e040fdb330 chore: merge upstream CLIProxyAPI v6.9.15
# Conflicts:
#	.gitignore
#	README.md
#	README_CN.md
#	README_JA.md
#	cmd/server/main.go
2026-04-05 13:55:36 +07:00
Luis Pater ada8e2905e feat(api): enhance proxy resolution for API key-based auth
Added comprehensive support for resolving proxy URLs from configuration based on API key and provider attributes. Introduced new helper functions and extended the test suite to validate fallback mechanisms and compatibility cases.
2026-04-05 01:56:34 +08:00
Luis Pater 4ba10531da feat(docs): add Poixe AI sponsorship details to README files
Added Poixe AI sponsorship information, including referral bonuses and platform capabilities, to README files in English, Japanese, and Chinese. Updated assets to include Poixe AI logo.
2026-04-05 01:20:50 +08:00
RCD 81929c06da fix(security): remove unverified management asset fallback (#12) 2026-04-04 21:25:56 +07:00
Luis Pater 3774b56e9f feat(misc): add background updater for Antigravity version caching
Introduce `StartAntigravityVersionUpdater` to periodically refresh the cached Antigravity version using a non-blocking background process. Updated main server flow to initialize the updater.
2026-04-04 22:09:11 +08:00
Luis Pater c2d4137fb9 feat(executor): enhance Qwen system message handling with strict injection and merging rules
Closes: #2537
2026-04-04 21:51:02 +08:00
Luis Pater 2ee938acaf Merge pull request #2535 from rensumo/main
feat: 动态获取 Antigravity User-Agent 版本号
2026-04-04 21:00:47 +08:00
rensumo 8d5e470e1f feat: dynamically fetch antigravity UA version from releases API
Fetch the latest version from the antigravity auto-updater releases
endpoint and cache it for 6 hours. Falls back to 1.21.9 if the API
is unreachable or returns unexpected data.
2026-04-04 14:52:59 +08:00
James 65e9e892a4 Fix missing response.completed.usage for late-usage OpenAI-compatible streams 2026-04-04 05:58:04 +00:00
Luis Pater 8430b28cfa Merge pull request #2526 from rensumo/main
feat: 升级反重力 (antigravity) UA 版本为 1.21.9
2026-04-04 11:32:16 +08:00
rensumo f3ab8f4bc5 chore: update antigravity UA version to 1.21.9 2026-04-04 07:35:08 +08:00
Luis Pater 0e4f189c2e Merge pull request #1302 from dinhkarate/feat(vertex)/add-prefix-field
Feat(vertex): add prefix field
2026-04-04 04:17:12 +08:00
Luis Pater 754b126944 fix(executor): remove commented-out code in QwenExecutor 2026-04-04 02:14:48 +08:00
Luis Pater ae37ccffbf Merge pull request #2520 from Arronlong/main
fix:qwen invalid_parameter_error
2026-04-04 02:13:09 +08:00
RCD d548446ce8 docs(readme): document default config bootstrap 2026-04-04 00:32:55 +07:00
RCD 852642d5ee fix(startup): bootstrap packaged default config 2026-04-04 00:32:36 +07:00
RCD 371bfc3528 docs: record upstream v6.9.12 sync 2026-04-04 00:22:10 +07:00
RCD 9cf230cc40 chore: merge upstream CLIProxyAPI v6.9.12 2026-04-04 00:21:26 +07:00
Luis Pater 42c062bb5b Merge pull request #2509 from adamhelfgott/fix-claude-thinking-temperature
Normalize Claude temperature when thinking is enabled
2026-04-03 23:55:50 +08:00
Luis Pater f389667ec3 Merge pull request #2513 from lonr-6/codex/fix-ws-custom-tool-repair-v2
fix: repair responses websocket custom tool call pairing
2026-04-03 23:45:38 +08:00
Arronlong 29dba0399b Comment out system message check in Qwen executor
fix qwen invalid_parameter_error
2026-04-03 23:07:33 +08:00
Luis Pater a824e7cd0b feat(models): add GPT-5.3, GPT-5.4, and GPT-5.4-mini with enhanced "thinking" levels 2026-04-03 23:05:10 +08:00
Luis Pater adb580b344 feat(security): add configuration to toggle Gemini CLI endpoint access
Closes: #2445
2026-04-03 21:46:49 +08:00
Luis Pater 06405f2129 fix(security): enforce stricter localhost validation for GeminiCLIAPIHandler
Closes: #2445
2026-04-03 21:22:03 +08:00
Kai Wang d1fd2c4ad4 fix: repair websocket custom tool calls 2026-04-03 17:11:44 +08:00
Kai Wang b6c6379bfa fix: repair websocket custom tool calls 2026-04-03 17:11:42 +08:00
Kai Wang 8f0e66b72e fix: repair websocket custom tool calls 2026-04-03 17:11:41 +08:00
Adam Helfgott f63cf6ff7a Normalize Claude temperature for thinking 2026-04-03 03:45:51 -04:00
RCD de8d8cc432 chore: restore executor compatibility wrappers 2026-04-03 13:16:37 +07:00
RCD 3d853cb295 chore: merge upstream CLIProxyAPI v6.9.10 2026-04-03 13:13:41 +07:00
Luis Pater d2419ed49d feat(executor): ensure default system message in QwenExecutor payload 2026-04-03 11:18:48 +08:00
mpfo0106 9b5ce8c64f Keep Claude builtin helpers aligned with the shared helper layout
The review asked for the builtin tool registry helper to live with the rest
of executor support utilities. This moves the registry code into the helps
package, exports the minimal surface executor needs, and keeps behavior tests
with the executor while leaving registry-focused checks with the helper.

Constraint: Requested layout keeps executor helper utilities centralized under internal/runtime/executor/helps
Rejected: Keep the files in executor and reply with rationale | conflicts with requested package organization
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: Keep executor behavior tests near applyClaudeToolPrefix and keep pure registry tests in helps
Tested: go test ./internal/runtime/executor/helps ./internal/runtime/executor -run 'Claude|Builtin|Tool'; go test ./test/...; go test ./...
Not-tested: End-to-end Claude Code direct-connect/session runtime behavior
2026-04-03 00:13:02 +09:00
Duong M. CUONG 058793c73a feat(gitstore): honor configured branch and follow live remote default 2026-04-02 14:44:44 +00:00
Luis Pater ab9ebea592 Merge PR #2474
# Conflicts:
#	internal/api/modules/amp/response_rewriter.go
#	internal/api/modules/amp/response_rewriter_test.go
2026-04-02 22:31:12 +08:00
Luis Pater 7ee37ee4b9 feat: add /healthz endpoint and test coverage for health check
Closes: #2493
2026-04-02 21:56:27 +08:00
Luis Pater 3171d524f0 docs: fix duplicated ProxyPal entry in README files 2026-04-02 21:22:40 +08:00
Luis Pater 3e78a8d500 Merge branch 'main' into dev 2026-04-02 21:21:26 +08:00
Luis Pater fcba912cc4 Merge pull request #2492 from davidwushi1145/main
fix(responses): reassemble split SSE event/data frames before streaming
2026-04-02 21:20:31 +08:00
Luis Pater 7170eeea5f Merge pull request #2454 from buddingnewinsights/add-proxypal-to-readme
docs: add ProxyPal to "Who is with us?" section
2026-04-02 21:18:22 +08:00
Luis Pater e3eb048c7a Merge pull request #2489 from Soein/upstream-pr
fix: 增强 Claude 反代检测对抗能力
2026-04-02 21:16:58 +08:00
Luis Pater a59e92435b Merge pull request #2490 from router-for-me/logs
Refactor websocket logging and error handling
2026-04-02 20:47:31 +08:00
davidwushi1145 108895fc04 Harden Responses SSE framing against partial chunk boundaries
Follow-up review found two real framing hazards in the handler-layer
framer: it could flush a partial `data:` payload before the JSON was
complete, and it could inject an extra newline before chunks that
already began with `\n`/`\r\n`. This commit tightens the framer so it
only emits undelimited events when the buffered `data:` payload is
already valid JSON (or `[DONE]`), skips newline injection for chunks
that already start with a line break, and avoids the heavier
`bytes.Split` path while scanning SSE fields.

The regression suite now covers split `data:` payload chunks,
newline-prefixed chunks, and dropping incomplete trailing data on
flush, so the original Responses fix remains intact while the review
concerns are explicitly locked down.

Constraint: Keep the follow-up limited to handler-layer framing and tests
Rejected: Ignore the review and rely on current executor chunk shapes | leaves partial data payload corruption possible
Rejected: Build a fully generic SSE parser | wider change than needed for the identified risks
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: Do not emit undelimited Responses SSE events unless buffered `data:` content is already complete and valid
Tested: /tmp/go1.26.1/go/bin/go test ./sdk/api/handlers/openai -count=1
Tested: /tmp/go1.26.1/go/bin/go test ./sdk/api/handlers -count=1
Tested: /tmp/go1.26.1/go/bin/go vet ./sdk/api/handlers/...
Not-tested: Full repository test suite outside sdk/api/handlers packages
2026-04-02 20:39:49 +08:00
davidwushi1145 abc293c642 Prevent malformed Responses SSE frames from breaking stream clients
Line-oriented upstream executors can emit `event:` and `data:` as
separate chunks, but the Responses handler had started terminating
each incoming chunk as a full SSE event. That split `response.created`
into an empty event plus a later data block, which broke downstream
clients like OpenClaw.

This keeps the fix in the handler layer: a small stateful framer now
buffers standalone `event:` lines until the matching `data:` arrives,
preserves already-framed events, and ignores delimiter-only leftovers.
The regression suite now covers split event/data framing, full-event
passthrough, terminal errors, and the bootstrap path that forwards
line-oriented openai-response streams from non-Codex executors too.

Constraint: Keep the fix localized to Responses handler framing instead of patching every executor
Rejected: Revert to v6.9.7 chunk writing | would reintroduce data-only framing regressions
Rejected: Patch each line-oriented executor separately | duplicates fragile SSE assembly logic
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: Do not assume incoming Responses stream chunks are already complete SSE events; preserve handler-layer reassembly for split `event:`/`data:` inputs
Tested: /tmp/go1.26.1/go/bin/go test ./sdk/api/handlers/openai -count=1
Tested: /tmp/go1.26.1/go/bin/go test ./sdk/api/handlers -count=1
Tested: /tmp/go1.26.1/go test ./sdk/api/handlers/... -count=1
Tested: /tmp/go1.26.1/go/bin/go vet ./sdk/api/handlers/...
Tested: Temporary patched server on 127.0.0.1:18317 -> /v1/models 200, /v1/responses non-stream 200, /v1/responses stream emitted combined `event:` + `data:` frames
Not-tested: Full repository test suite outside sdk/api/handlers packages
2026-04-02 20:26:42 +08:00
mpfo0106 da3a498a28 Keep Claude Code compatibility work low-risk and reviewable
This change stops short of broader Claude Code runtime alignment and instead
hardens two safe edges: builtin tool prefix handling and source-informed
sentinel coverage for future drift checks.

Constraint: Must preserve existing default behavior for current users
Rejected: Implement control-plane/session alignment now | too much runtime risk for a first slice
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: Treat the new fixtures as compatibility sentinels, not a full Claude Code schema contract
Tested: go test ./test/...; go test ./sdk/translator/...; go test ./internal/runtime/executor -run 'Claude|Builtin|Tool'; go test ./...
Not-tested: End-to-end Claude Code direct-connect/session runtime behavior
2026-04-02 20:35:39 +09:00
pzy bb44671845 fix: 修复反代检测对抗的 3 个问题
- computeFingerprint 使用 rune 索引替代字节索引,修复多字节字符指纹不匹配
- utls Chrome TLS 指纹仅对 Anthropic 官方域名生效,自定义 base_url 走标准 transport
- IPv6 地址使用 net.JoinHostPort 正确拼接端口
2026-04-02 19:12:55 +08:00
Luis Pater 09e480036a feat(auth): add support for managing custom headers in auth files
Closes #2457
2026-04-02 19:11:09 +08:00
pzy 249f969110 fix: Claude API 请求使用 utls Chrome TLS 指纹
Claude executor 的 API 请求之前使用 Go 标准库 crypto/tls,JA3 指纹
与真实 Claude Code(Bun/BoringSSL)不匹配,可被 Cloudflare 识别。

- 新增 helps/utls_client.go,封装 utls Chrome 指纹 + HTTP/2 + 代理支持
- Claude executor 的 4 处 NewProxyAwareHTTPClient 替换为 NewUtlsHTTPClient
- 其他 executor(Gemini/Codex/iFlow 等)不受影响,仍用标准 TLS
- 非 HTTPS 请求自动回退到标准 transport
2026-04-02 19:09:56 +08:00
hkfires 4f8acec2d8 refactor(logging): centralize websocket handshake recording 2026-04-02 18:39:32 +08:00
hkfires 34339f61ee Refactor websocket logging and error handling
- Introduced new logging functions for websocket requests, handshakes, errors, and responses in `logging_helpers.go`.
- Updated `CodexWebsocketsExecutor` to utilize the new logging functions for improved clarity and consistency in websocket operations.
- Modified the handling of websocket upgrade rejections to log relevant metadata.
- Changed the request body key to a timeline body key in `openai_responses_websocket.go` to better reflect its purpose.
- Enhanced tests to verify the correct logging of websocket events and responses, including disconnect events and error handling scenarios.
2026-04-02 17:30:51 +08:00
pzy 4045378cb4 fix: 增强 Claude 反代检测对抗能力
基于 Claude Code v2.1.88 源码分析,修复多个可被 Anthropic 检测的差距:

- 实现消息指纹算法(SHA256 盐值 + 字符索引),替代随机 buildHash
- billing header cc_version 从设备 profile 动态取版本号,不再硬编码
- billing header cc_entrypoint 从客户端 UA 解析,支持 cli/vscode/local-agent
- billing header 新增 cc_workload 支持(通过 X-CPA-Claude-Workload 头传入)
- 新增 X-Claude-Code-Session-Id 头(每 apiKey 缓存 UUID,TTL=1h)
- 新增 x-client-request-id 头(仅 api.anthropic.com,每请求 UUID)
- 补全 4 个缺失的 beta flags(structured-outputs/fast-mode/redact-thinking/token-efficient-tools)
- OAuth scope 对齐 Claude Code 2.1.88(移除 org:create_api_key,添加 sessions/mcp/file_upload)
- Anthropic-Dangerous-Direct-Browser-Access 仅在 API key 模式发送
- 响应头网关指纹清洗(剥离 litellm/helicone/portkey/cloudflare/kong/braintrust 前缀头)
2026-04-02 15:55:22 +08:00
Luis Pater 4f99bc54f1 test: update codex header expectations 2026-04-02 11:19:37 +08:00
Luis Pater 913f4a9c5f test: fix executor tests after helpers refactor 2026-04-02 11:12:30 +08:00
Luis Pater 25d1c18a3f fix: scope experimental cch signing to billing header 2026-04-02 11:03:11 +08:00
Luis Pater d09dd4d0b2 Merge commit '15c2f274ea690c9a7c9db22f9f454af869db5375' into dev 2026-04-02 10:59:54 +08:00
Luis Pater 474fb042da Merge pull request #2476 from router-for-me/cherry-pick/pr-2438-to-dev
Cherry-pick PR #2438 onto dev
2026-04-02 10:36:50 +08:00
Michael 8435c3d7be feat(tui): show time in usage details 2026-04-02 10:35:13 +08:00
Luis Pater e783d0a62e Merge pull request #2441 from MonsterQiu/issue-2421-alias-before-suspension
fix(auth): resolve oauth aliases before suspension checks
2026-04-02 10:27:39 +08:00
Luis Pater b05f575e9b Merge pull request #2444 from 0oAstro/fix/codex-nonstream-finish-reason-tool-calls
fix(codex): set finish_reason to "tool_calls" in non-streaming response when tool calls are present
2026-04-02 10:01:25 +08:00
Aikins Laryea f5e9f01811 test(amp): update tests to expect thinking blocks to pass through during streaming 2026-04-01 20:35:23 +00:00
Aikins Laryea ff7dbb5867 test(amp): update tests to expect thinking blocks to pass through during streaming 2026-04-01 20:21:39 +00:00
Aikins Laryea e34b2b4f1d fix(gemini): clean tool schemas and eager_input_streaming
delegate schema sanitization to util.CleanJSONSchemaForGemini and drop the top-level eager_input_streaming key to prevent validation errors when sending claude tools to the gemini api
2026-04-01 19:49:38 +00:00
edlsh 15c2f274ea fix: preserve cloak config defaults when mode omitted 2026-04-01 13:20:11 -04:00
edlsh 37249339ac feat: add opt-in experimental Claude cch signing 2026-04-01 13:03:17 -04:00
Luis Pater c422d16beb Merge pull request #2398 from 7RPH/fix/responses-sse-framing
fix: preserve SSE event boundaries for Responses streams
2026-04-02 00:46:51 +08:00
Luis Pater 66cd50f603 Merge pull request #2468 from router-for-me/ip
fix(openai): improve client IP retrieval in websocket handler
2026-04-02 00:03:35 +08:00
hkfires caa529c282 fix(openai): improve client IP retrieval in websocket handler 2026-04-01 20:16:01 +08:00
hkfires 51a4379bf4 refactor(openai): remove websocket body log truncation limit 2026-04-01 18:11:43 +08:00
Luis Pater acf98ed10e fix(openai): add session reference counter and cache lifecycle management for websocket tools 2026-04-01 17:28:50 +08:00
Luis Pater d1c07a091e fix(openai): add websocket tool call repair with caching and tests to improve transcript consistency 2026-04-01 17:16:49 +08:00
Luis Pater 105a21548f fix(codex): centralize session management with global store and add tests for executor session lifecycle 2026-04-01 13:17:10 +08:00
Luis Pater 1734aa1664 fix(codex): prioritize websocket-enabled credentials across priority tiers in scheduler logic 2026-04-01 12:51:12 +08:00
Luis Pater ca11b236a7 refactor(runtime, openai): simplify header management and remove redundant websocket logging logic 2026-04-01 11:57:31 +08:00
huynhgiabuu 6fdff8227d docs: add ProxyPal to 'Who is with us?' section
Add ProxyPal (https://github.com/buddingnewinsights/proxypal) to the
community projects list in all three README files (EN, CN, JA).
Placed after CCS, restoring its original position.

ProxyPal is a cross-platform desktop app (macOS, Windows, Linux) that
wraps CLIProxyAPI with a native GUI, supporting multiple AI providers,
usage analytics, request monitoring, and auto-configuration for popular
coding tools.

Closes #2420
2026-04-01 10:23:22 +07:00
Luis Pater 330e12d3c2 fix(codex): conditionally set Session_id header for Mac OS user agents and clean up redundant logic 2026-04-01 11:11:45 +08:00
RCD 8fc58e0806 chore: merge upstream CLIProxyAPI v6.9.7 2026-04-01 03:36:56 +07:00
Luis Pater b468ca79c3 Merge branch 'dev' of github.com:router-for-me/CLIProxyAPI into dev 2026-04-01 03:09:03 +08:00
Luis Pater d2c7e4e96a refactor(runtime): move executor utilities to helps package and update references 2026-04-01 03:08:20 +08:00
Luis Pater 1c7003ff68 Merge pull request #2452 from Lucaszmv/fix-qwen-cli-v0.13.2
fix(qwen): update CLI simulation to v0.13.2 and adjust header casing
2026-04-01 02:44:27 +08:00
Lucaszmv 1b44364e78 fix(qwen): update CLI simulation to v0.13.2 2026-03-31 15:19:07 -03:00
0oAstro ec77f4a4f5 fix(codex): set finish_reason to tool_calls in non-streaming response when tool calls are present 2026-03-31 14:12:15 +05:30
MonsterQiu f611dd6e96 refactor(auth): dedupe route-aware model support checks 2026-03-31 15:42:25 +08:00
MonsterQiu 07b7c1a1e0 fix(auth): resolve oauth aliases before suspension checks 2026-03-31 14:27:14 +08:00
Luis Pater 51fd58d74f fix(codex): use normalizeCodexInstructions to set default instructions 2026-03-31 12:16:57 +08:00
Luis Pater faae9c2f7c Merge pull request #2422 from MonsterQiu/fix/codex-compact-instructions
fix(codex): add default instructions for /responses/compact
2026-03-31 12:14:20 +08:00
Luis Pater bc3a6e4646 Merge pull request #2434 from MonsterQiu/fix/codex-responses-null-instructions
fix(codex): normalize null instructions for /responses requests
2026-03-31 12:01:21 +08:00
Luis Pater b09b03e35e Merge pull request #2424 from possible055/fix/websocket-transcript-replacement
fix(openai): handle transcript replacement after websocket v2 compaction
2026-03-31 11:00:33 +08:00
Luis Pater 16231947e7 Merge pull request #2426 from xixiwenxuanhe/feature/antigravity-credits
feat(antigravity): add AI credits quota fallback
2026-03-31 10:51:40 +08:00
MonsterQiu 39b9a38fbc fix(codex): normalize null instructions across responses paths 2026-03-31 10:32:39 +08:00
MonsterQiu bd855abec9 fix(codex): normalize null instructions for responses requests 2026-03-31 10:29:02 +08:00
Luis Pater 7c3c2e9f64 Merge pull request #2417 from CharTyr/fix/amp-streaming-thinking-regression
fix(amp): 修复流式响应中 thinking block 被错误抑制导致的 TUI 空白回复
2026-03-31 10:12:13 +08:00
Luis Pater c10f8ae2e2 Fixed: #2420
docs(readme): remove ProxyPal section from all README translations
2026-03-31 07:23:02 +08:00
xixiwenxuanhe a0bf33eca6 fix(antigravity): preserve fallback and honor config gate 2026-03-31 00:14:05 +08:00
xixiwenxuanhe 88dd9c715d feat(antigravity): add AI credits quota fallback 2026-03-30 23:58:12 +08:00
apparition a3e21df814 fix(openai): avoid developer transcript resets
- Narrow websocket transcript replacement detection to assistant outputs and function calls
- Preserve existing merge behavior for follow-up developer messages without previous_response_id
- Add a regression test covering mid-session developer message updates
2026-03-30 23:33:16 +08:00
MonsterQiu d3b94c9241 fix(codex): normalize null instructions for compact requests 2026-03-30 22:58:05 +08:00
apparition c1d7599829 fix(openai): handle transcript replacement after websocket compaction
- Add shouldReplaceWebsocketTranscript() to detect historical model output in input
- Add normalizeResponseTranscriptReplacement() for full transcript reset handling
- Prevent duplicate stale turn-state when clients replace local history post-compaction
- Avoid orphaned function_call items from incremental append on compact transcripts
- Add unit tests for transcript replacement detection and state reset behavior
2026-03-30 22:44:58 +08:00
MonsterQiu d11936f292 fix(codex): add default instructions for /responses/compact 2026-03-30 22:44:46 +08:00
Luis Pater 17363edf25 fix(auth): skip downtime for request-scoped 404 errors in model state management 2026-03-30 22:22:42 +08:00
CharTyr 279cbbbb8a fix(amp): don't suppress thinking blocks in streaming mode
Reverts the streaming thinking suppression introduced in b15453c.
rewriteStreamEvent should only inject signatures and rewrite model
names — suppressing thinking blocks in streaming mode breaks SSE
index alignment and causes the Amp TUI to render empty responses
on the second message onward (especially with model-mapped
non-Claude providers like GPT-5.4).

Non-streaming responses still suppress thinking when tool_use is
present via rewriteModelInResponse.
2026-03-30 20:09:32 +08:00
daniel 91387ca247 refactor(gemini-cli): simplify redundant if/else in project ID assignment
Both branches assign finalProjectID = responseProjectID, so move the
assignment outside the conditional and keep only the logging inside.
2026-03-30 20:07:02 +08:00
Luis Pater 486cd4c343 Merge pull request #2409 from sususu98/fix/tool-use-pairing-break
fix(antigravity): reorder model parts to prevent tool_use↔tool_result pairing breakage
2026-03-30 16:59:46 +08:00
sususu98 25feceb783 fix(antigravity): reorder model parts to prevent tool_use↔tool_result pairing breakage
When a Claude assistant message contains [text, tool_use, text], the
Antigravity API internally splits the model message at functionCall
boundaries, creating an extra assistant turn between tool_use and the
following tool_result. Claude then rejects with:

  tool_use ids were found without tool_result blocks immediately after

Fix: extend the existing 2-way part reordering (thinking-first) to a
3-way partition: thinking → regular → functionCall. This ensures
functionCall parts are always last, so Antigravity's split cannot
insert an extra assistant turn before the user's tool_result.

Fixes #989
2026-03-30 15:09:33 +08:00
Luis Pater d26752250d Merge pull request #2403 from CharTyr/clean-pr
fix(amp): 修复Amp CLI 集成 缺失/无效 signature 导致的 TUI 崩溃与上游 400 问题
2026-03-30 12:54:15 +08:00
CharTyr b15453c369 fix(amp): address PR review - stream thinking suppression, SSE detection, test init
- Call suppressAmpThinking in rewriteStreamEvent for streaming path
- Handle nil return from suppressAmpThinking to skip suppressed events
- Narrow looksLikeSSEChunk to line-prefix detection (HasPrefix vs Contains)
- Initialize suppressedContentBlock map in test
2026-03-30 00:42:04 -04:00
CharTyr 04ba8c8bc3 feat(amp): sanitize signatures and handle stream suppression for Amp compatibility 2026-03-29 22:23:18 -04:00
danielandClaude Sonnet 4.6 fccfb162b4 fix(gemini-cli): use backend project ID from onboarding response
- Simplify project ID selection to always use the backend project ID returned by Gemini onboarding
- Update Gemini CLI version from 0.31.0 to 0.34.0
- Add 'terminal' to User-Agent string for better client identification

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-29 23:52:01 +08:00
Luis Pater 6570692291 Merge pull request #2400 from router-for-me/revert-2374-codex-cache-clean
Revert "fix(codex): restore prompt cache continuity for Codex requests"
2026-03-29 22:19:39 +08:00
trph f73d55ddaa fix: simplify responses SSE suffix handling 2026-03-29 22:19:25 +08:00
Luis Pater 13aa5b3375 Revert "fix(codex): restore prompt cache continuity for Codex requests" 2026-03-29 22:18:14 +08:00
trph 0fcc02fbea fix: tighten responses SSE review follow-up 2026-03-29 22:10:28 +08:00
trph c03883ccf0 fix: address responses SSE review feedback 2026-03-29 22:00:46 +08:00
trph 134a9eac9d fix: preserve SSE event boundaries for Responses streams 2026-03-29 17:23:16 +08:00
Luis Pater 6d8de0ade4 feat(auth): implement weighted provider rotation for improved scheduling fairness 2026-03-29 13:49:01 +08:00
Luis Pater 1587ff5e74 Merge pull request #2389 from router-for-me/claude
fix(claude): add default max_tokens for models
2026-03-29 13:03:20 +08:00
hkfires f033d3a6df fix(claude): enhance ensureModelMaxTokens to use registered max_completion_tokens and fallback to default 2026-03-29 13:00:43 +08:00
hkfires 145e0e0b5d fix(claude): add default max_tokens for models 2026-03-29 12:46:00 +08:00
Ravi Tharuma 5fc2bd393e fix: retain codex thinking signature until item done 2026-03-28 14:41:25 +01:00
Ravi Tharuma 66eb12294a fix: clear stale thinking signature when no block is open 2026-03-28 14:08:31 +01:00
Ravi Tharuma 73b22ec29b fix: omit empty signature field from thinking blocks
Emit signature only when non-empty in both streaming content_block_start
and non-streaming thinking blocks. Avoids turning 'missing signature'
into 'empty/invalid signature' which Claude clients may reject.
2026-03-28 14:08:31 +01:00
Ravi Tharuma c31ae2f3b5 fix: retain previously captured thinking signature on new summary part 2026-03-28 14:08:31 +01:00
Ravi Tharuma 76b53d6b5b fix: finalize pending thinking block before next summary part 2026-03-28 14:08:31 +01:00
Ravi Tharuma a34dfed378 fix: preserve Claude thinking signatures in Codex translator 2026-03-28 14:08:31 +01:00
dinhkarateandClaude Opus 4.6 36efcc6e28 fix(vertex): include prefix in auth filename and validate at import
Address two blocking issues from PR review:
- Auth file now named vertex-{prefix}-{project}.json so importing the
  same project with different prefixes no longer overwrites credentials
- Prefix containing "/" is rejected at import time instead of being
  silently ignored at runtime
- Add prefix to in-memory metadata map for consistency

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-17 15:06:04 +07:00
Pham Quang Dinh a337ecf35c Merge branch 'router-for-me:main' into feat(vertex)/add-prefix-field 2026-03-17 11:48:40 +07:00
destinoantagonista-wq e08f68ed7c chore(auth): drop reconcile test file from pr 2026-03-14 14:41:26 +00:00
destinoantagonista-wq f09ed25fd3 fix(auth): tighten registry model reconciliation 2026-03-14 14:40:06 +00:00
destinoantagonista-wq e166e56249 Reconcile registry model states on auth changes
Add Manager.ReconcileRegistryModelStates to clear stale per-model runtime failures for models currently registered in the global model registry. The method finds models supported for an auth, resets non-clean ModelState entries, updates aggregated availability, persists changes, and pushes a snapshot to the scheduler. Introduce modelStateIsClean helper to determine when a model state needs resetting. Call ReconcileRegistryModelStates from Service paths that register/refresh models (applyCoreAuthAddOrUpdate and refreshModelRegistrationForAuth) to keep the scheduler and global registry aligned after model re-registration.
2026-03-13 19:41:49 +00:00
Blue-B 5f58248016 fix(claude): clamp max_tokens to model limit in normalizeClaudeBudget
When adjustedBudget < minBudget, the previous fix blindly set
max_tokens = budgetTokens+1 which could exceed MaxCompletionTokens.

Now: cap max_tokens at MaxCompletionTokens, recalculate budget, and
disable thinking entirely if constraints are unsatisfiable.

Add unit tests covering raise, clamp, disable, and no-op scenarios.
2026-03-09 22:10:30 +09:00
Blue-B 07d6689d87 fix(claude): add interleaved-thinking beta header, AMP gzip error decoding, normalizeClaudeBudget max_tokens
1. Always include interleaved-thinking-2025-05-14 beta header so that
   thinking blocks are returned correctly for all Claude models.

2. Remove status-code guard in AMP reverse proxy ModifyResponse so that
   error responses (4xx/5xx) with hidden gzip encoding are decoded
   properly — prevents garbled error messages reaching the client.

3. In normalizeClaudeBudget, when the adjusted budget falls below the
   model minimum, set max_tokens = budgetTokens+1 instead of leaving
   the request unchanged (which causes a 400 from the API).
2026-03-07 21:31:10 +09:00
dinhkarate 14cb2b95c6 feat(vertex): add --vertex-import-prefix flag for model namespacing 2026-01-29 13:32:38 +07:00
dinhkarate fdeef48498 feat(vertex): Add Prefix field to VertexCredentialStorage for per-file model namespacing 2026-01-29 13:32:38 +07:00
582 changed files with 47929 additions and 7173 deletions
+81
View File
@@ -0,0 +1,81 @@
name: agents-md-guard
on:
pull_request:
types:
- opened
- synchronize
- reopened
permissions:
contents: read
issues: write
pull-requests: write
jobs:
close-when-agents-md-changed:
runs-on: ubuntu-latest
steps:
- name: Detect AGENTS.md changes and close PR
uses: actions/github-script@v7
with:
script: |
const prNumber = context.payload.pull_request.number;
const { owner, repo } = context.repo;
const files = await github.paginate(github.rest.pulls.listFiles, {
owner,
repo,
pull_number: prNumber,
per_page: 100,
});
const touchesAgentsMd = (path) =>
typeof path === "string" &&
(path === "AGENTS.md" || path.endsWith("/AGENTS.md"));
const touched = files.filter(
(f) => touchesAgentsMd(f.filename) || touchesAgentsMd(f.previous_filename),
);
if (touched.length === 0) {
core.info("No AGENTS.md changes detected.");
return;
}
const changedList = touched
.map((f) =>
f.previous_filename && f.previous_filename !== f.filename
? `- ${f.previous_filename} -> ${f.filename}`
: `- ${f.filename}`,
)
.join("\n");
const body = [
"This repository does not allow modifying `AGENTS.md` in pull requests.",
"",
"Detected changes:",
changedList,
"",
"Please revert these changes and open a new PR without touching `AGENTS.md`.",
].join("\n");
try {
await github.rest.issues.createComment({
owner,
repo,
issue_number: prNumber,
body,
});
} catch (error) {
core.warning(`Failed to comment on PR #${prNumber}: ${error.message}`);
}
await github.rest.pulls.update({
owner,
repo,
pull_number: prNumber,
state: "closed",
});
core.setFailed("PR modifies AGENTS.md");
@@ -0,0 +1,73 @@
name: auto-retarget-main-pr-to-dev
on:
pull_request:
types:
- opened
- reopened
- edited
branches:
- main
permissions:
contents: read
issues: write
pull-requests: write
jobs:
retarget:
if: github.actor != 'github-actions[bot]'
runs-on: ubuntu-latest
steps:
- name: Retarget PR base to dev
uses: actions/github-script@v7
with:
script: |
const pr = context.payload.pull_request;
const prNumber = pr.number;
const { owner, repo } = context.repo;
const baseRef = pr.base?.ref;
const headRef = pr.head?.ref;
const desiredBase = "dev";
if (baseRef !== "main") {
core.info(`PR #${prNumber} base is ${baseRef}; nothing to do.`);
return;
}
if (headRef === desiredBase) {
core.info(`PR #${prNumber} is ${desiredBase} -> main; skipping retarget.`);
return;
}
core.info(`Retargeting PR #${prNumber} base from ${baseRef} to ${desiredBase}.`);
try {
await github.rest.pulls.update({
owner,
repo,
pull_number: prNumber,
base: desiredBase,
});
} catch (error) {
core.setFailed(`Failed to retarget PR #${prNumber} to ${desiredBase}: ${error.message}`);
return;
}
const body = [
`This pull request targeted \`${baseRef}\`.`,
"",
`The base branch has been automatically changed to \`${desiredBase}\`.`,
].join("\n");
try {
await github.rest.issues.createComment({
owner,
repo,
issue_number: prNumber,
body,
});
} catch (error) {
core.warning(`Failed to comment on PR #${prNumber}: ${error.message}`);
}
-1
View File
@@ -22,7 +22,6 @@ jobs:
- uses: actions/setup-go@v4
with:
go-version: '>=1.26.0'
cache: true
- name: Generate Build Metadata
run: |
VERSION=$(git describe --tags --always --dirty)
+33
View File
@@ -0,0 +1,33 @@
name: rust-replatform
on:
pull_request:
paths:
- "rust/**"
- "docs/rust-replatform.md"
- ".github/workflows/rust-replatform.yml"
push:
branches:
- rust-replatform-v2
paths:
- "rust/**"
- "docs/rust-replatform.md"
- ".github/workflows/rust-replatform.yml"
workflow_dispatch:
permissions:
contents: read
jobs:
test:
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v4
- name: Set up Rust
uses: dtolnay/rust-toolchain@stable
- name: Test
working-directory: rust
run: cargo test -p proxypilot-rs
+11 -2
View File
@@ -1,6 +1,6 @@
# Binaries
cli-proxy-api
cliproxy
/cli-proxy-api
/cliproxy
proxypilot-linux
proxypilot-darwin
proxypilot-windows
@@ -55,6 +55,7 @@ GEMINI.md
# Tooling metadata
.vscode/*
.worktrees/
.codex/*
.claude/*
.gemini/*
@@ -63,6 +64,7 @@ GEMINI.md
.agents/*
.opencode/*
.idea/*
.beads/*
.bmad/*
_bmad/*
_bmad-output/*
@@ -83,6 +85,7 @@ tui-ttyd.png
# macOS
.DS_Store
._*
.gocache/
# Windows artifacts
nul
@@ -118,3 +121,9 @@ config-debug.yaml
test_results.txt
test_results_full.txt
tmpclaude-*
# Opencode
.beads/
.opencode/
.cli-proxy-api/
.venv/
+2
View File
@@ -27,6 +27,8 @@ builds:
archives:
- id: "cli-proxy-api-plus"
format: tar.gz
name_template: >-
{{ .ProjectName }}_{{ .Version }}_{{ .Os }}_{{- if eq .Arch "arm64" -}}aarch64{{- else -}}{{ .Arch }}{{- end -}}
format_overrides:
- goos: windows
format: zip
+1
View File
@@ -0,0 +1 @@
15ac7fb9324095330e60f522147b8a8e81f16ab5
+1
View File
@@ -0,0 +1 @@
v7.0.2
+58
View File
@@ -0,0 +1,58 @@
# AGENTS.md
Go 1.26+ proxy server providing OpenAI/Gemini/Claude/Codex compatible APIs with OAuth and round-robin load balancing.
## Repository
- GitHub: https://github.com/router-for-me/CLIProxyAPI
## Commands
```bash
gofmt -w . # Format (required after Go changes)
go build -o cli-proxy-api ./cmd/server # Build
go run ./cmd/server # Run dev server
go test ./... # Run all tests
go test -v -run TestName ./path/to/pkg # Run single test
go build -o test-output ./cmd/server && rm test-output # Verify compile (REQUIRED after changes)
```
- Common flags: `--config <path>`, `--tui`, `--standalone`, `--local-model`, `--no-browser`, `--oauth-callback-port <port>`
## Config
- Default config: `config.yaml` (template: `config.example.yaml`)
- `.env` is auto-loaded from the working directory
- Auth material defaults under `auths/`
- Storage backends: file-based default; optional Postgres/git/object store (`PGSTORE_*`, `GITSTORE_*`, `OBJECTSTORE_*`)
## Architecture
- `cmd/server/` — Server entrypoint
- `internal/api/` — Gin HTTP API (routes, middleware, modules)
- `internal/api/modules/amp/` — Amp integration (Amp-style routes + reverse proxy)
- `internal/thinking/` — Main thinking/reasoning pipeline. `ApplyThinking()` (apply.go) parses suffixes (`suffix.go`, suffix overrides body), normalizes config to canonical `ThinkingConfig` (`types.go`), normalizes and validates centrally (`validate.go`/`convert.go`), then applies provider-specific output via `ProviderApplier`. Do not break this "canonical representation → per-provider translation" architecture.
- `internal/runtime/executor/` — Per-provider runtime executors (incl. Codex WebSocket)
- `internal/translator/` — Provider protocol translators (and shared `common`)
- `internal/registry/` — Model registry + remote updater (`StartModelsUpdater`); `--local-model` disables remote updates
- `internal/store/` — Storage implementations and secret resolution
- `internal/managementasset/` — Config snapshots and management assets
- `internal/cache/` — Request signature caching
- `internal/watcher/` — Config hot-reload and watchers
- `internal/wsrelay/` — WebSocket relay sessions
- `internal/usage/` — Usage and token accounting
- `internal/tui/` — Bubbletea terminal UI (`--tui`, `--standalone`)
- `sdk/cliproxy/` — Embeddable SDK entry (service/builder/watchers/pipeline)
- `test/` — Cross-module integration tests
## Code Conventions
- Keep changes small and simple (KISS)
- Comments in English only
- If editing code that already contains non-English comments, translate them to English (don’t add new non-English comments)
- For user-visible strings, keep the existing language used in that file/area
- New Markdown docs should be in English unless the file is explicitly language-specific (e.g. `README_CN.md`)
- As a rule, do not make standalone changes to `internal/translator/`. You may modify it only as part of broader changes elsewhere.
- If a task requires changing only `internal/translator/`, run `gh repo view --json viewerPermission -q .viewerPermission` to confirm you have `WRITE`, `MAINTAIN`, or `ADMIN`. If you do, you may proceed; otherwise, file a GitHub issue including the goal, rationale, and the intended implementation code, then stop further work.
- `internal/runtime/executor/` should contain executors and their unit tests only. Place any helper/supporting files under `internal/runtime/executor/helps/`.
- Follow `gofmt`; keep imports goimports-style; wrap errors with context where helpful
- Do not use `log.Fatal`/`log.Fatalf` (terminates the process); prefer returning errors and logging via logrus
- Shadowed variables: use method suffix (`errStart := server.Start()`)
- Wrap defer errors: `defer func() { if err := f.Close(); err != nil { log.Errorf(...) } }()`
- Use logrus structured logging; avoid leaking secrets/tokens in logs
- Avoid panics in HTTP handlers; prefer logged errors and meaningful HTTP status codes
- Timeouts are allowed only during credential acquisition; after an upstream connection is established, do not set timeouts for any subsequent network behavior. Intentional exceptions that must remain allowed are the Codex websocket liveness deadlines in `internal/runtime/executor/codex_websockets_executor.go`, the wsrelay session deadlines in `internal/wsrelay/session.go`, the management APICall timeout in `internal/api/handlers/management/api_tools.go`, and the `cmd/fetch_antigravity_models` utility timeouts
+2 -2
View File
@@ -14,13 +14,13 @@ ARG BUILD_DATE=unknown
RUN CGO_ENABLED=0 GOOS=linux go build -ldflags="-s -w -X 'main.Version=${VERSION}-plus' -X 'main.Commit=${COMMIT}' -X 'main.BuildDate=${BUILD_DATE}'" -o ./CLIProxyAPIPlus ./cmd/server/
FROM alpine:3.22.0
FROM alpine:3.23
RUN apk add --no-cache tzdata
RUN mkdir /ProxyPilot
COPY --from=builder ./app/CLIProxyAPIPlus /CLIProxyAPI/CLIProxyAPIPlus
COPY --from=builder ./app/CLIProxyAPIPlus /ProxyPilot/CLIProxyAPIPlus
COPY config.example.yaml /ProxyPilot/config.example.yaml
+22 -6
View File
@@ -67,6 +67,8 @@
| Zhipu AI | API Key | GLM-4.5, GLM-4.6, GLM-4.7 |
| Custom | API Key | Any OpenAI-compatible endpoint |
> **Account risk warning:** Claude/Anthropic OAuth, Google/Gemini OAuth, Gemini CLI OAuth, and Antigravity OAuth are unofficial compatibility paths. Providers may treat proxy, automation, shared-token, or third-party OAuth traffic as a policy violation or abnormal usage. Using these login flows can risk quota loss, account restriction, or account suspension. Prefer official API-key or officially supported developer access for important accounts.
---
## Installation
@@ -92,10 +94,11 @@ go build -o proxypilot ./cmd/server
### First Launch
1. Copy config: `cp config.example.yaml config.yaml`
2. Run: `./proxypilot`
3. Server starts on `http://localhost:8317`
4. Open dashboard: `http://localhost:8317/management.html`
1. Packaged installs create `config.yaml` automatically on first start when `config.example.yaml` is bundled next to the executable.
2. If you're running from source, copy config: `cp config.example.yaml config.yaml`
3. Run: `./proxypilot`
4. Server starts on `http://localhost:8317`
5. Open dashboard: `http://localhost:8317/management.html`
### Authentication
@@ -106,8 +109,8 @@ Run OAuth login for your provider:
./proxypilot --claude-login # Claude
./proxypilot --codex-login # OpenAI/Codex
./proxypilot --login # Gemini
./proxypilot --kiro-login # Kiro (Google OAuth)
./proxypilot --kiro-aws-login # Kiro (AWS Builder ID)
./proxypilot --kiro-login # Kiro (AWS Builder ID)
./proxypilot --kiro-aws-login # Kiro (AWS Builder ID device code)
./proxypilot --qwen-login # Qwen
./proxypilot --antigravity-login # Antigravity
@@ -121,6 +124,10 @@ Run OAuth login for your provider:
OAuth tokens are stored locally and auto-refreshed before expiry.
**High-risk OAuth providers:** `--claude-login`, `--login` / Gemini, Gemini CLI OAuth, and `--antigravity-login` may violate provider expectations or trigger anti-abuse/account-protection systems. Do not use production, billing-critical, school/work, or irreplaceable personal accounts with these flows unless you accept the account-risk tradeoff. Prefer official API keys where available.
Kiro Google/GitHub OAuth is not currently supported because AWS Cognito rejects third-party localhost callbacks. Use AWS Builder ID (`--kiro-login` / `--kiro-aws-login`) or import from the official Kiro IDE.
### Security Defaults (Auth + CORS)
- Proxy requests require API keys by default. To allow unauthenticated access (not recommended), set `allow-unauthenticated: true` in `config.yaml`.
@@ -250,6 +257,15 @@ prompt-cache:
---
## Ecosystem
### [CPA Usage Keeper](https://github.com/Willxup/cpa-usage-keeper)
Standalone persistence and visualization service for CLIProxyAPI, with periodic data sync, SQLite storage, aggregate APIs, and a built-in dashboard for usage and statistics.
> [!NOTE]
> If you developed a project based on CLIProxyAPI, please open a PR to add it to this list.
## Auto-Updates
ProxyPilot includes a built-in auto-update system that checks for new releases and allows one-click installation.
+1 -1
View File
@@ -1 +1 @@
v6.9.5
v7.0.2
Binary file not shown.

After

Width:  |  Height:  |  Size: 170 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 401 KiB

+6 -5
View File
@@ -25,10 +25,11 @@ import (
"strings"
"time"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
sdkauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/proxyutil"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v7/internal/misc"
sdkauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/proxyutil"
log "github.com/sirupsen/logrus"
"github.com/tidwall/gjson"
)
@@ -188,7 +189,7 @@ func fetchModels(ctx context.Context, auth *coreauth.Auth) []modelEntry {
httpReq.Close = true
httpReq.Header.Set("Content-Type", "application/json")
httpReq.Header.Set("Authorization", "Bearer "+accessToken)
httpReq.Header.Set("User-Agent", "antigravity/1.19.6 darwin/arm64")
httpReq.Header.Set("User-Agent", misc.AntigravityUserAgent())
httpClient := &http.Client{Timeout: 30 * time.Second}
if transport, _, errProxy := proxyutil.BuildHTTPTransport(auth.ProxyURL); errProxy == nil && transport != nil {
+3 -3
View File
@@ -9,9 +9,9 @@ import (
"sync"
"time"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v6/internal/desktopctl"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/desktopctl"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy"
log "github.com/sirupsen/logrus"
)
+14 -14
View File
@@ -22,20 +22,20 @@ import (
"github.com/getlantern/systray"
"github.com/jchv/go-webview2"
configaccess "github.com/router-for-me/CLIProxyAPI/v6/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v6/internal/api/middleware"
"github.com/router-for-me/CLIProxyAPI/v6/internal/cmd"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v6/internal/desktopctl"
"github.com/router-for-me/CLIProxyAPI/v6/internal/integrations"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
_ "github.com/router-for-me/CLIProxyAPI/v6/internal/translator"
"github.com/router-for-me/CLIProxyAPI/v6/internal/trayicon"
"github.com/router-for-me/CLIProxyAPI/v6/internal/updates"
"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
"github.com/router-for-me/CLIProxyAPI/v6/internal/winutil"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
configaccess "github.com/router-for-me/CLIProxyAPI/v7/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v7/internal/api/middleware"
"github.com/router-for-me/CLIProxyAPI/v7/internal/cmd"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/desktopctl"
"github.com/router-for-me/CLIProxyAPI/v7/internal/integrations"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
_ "github.com/router-for-me/CLIProxyAPI/v7/internal/translator"
"github.com/router-for-me/CLIProxyAPI/v7/internal/trayicon"
"github.com/router-for-me/CLIProxyAPI/v7/internal/updates"
"github.com/router-for-me/CLIProxyAPI/v7/internal/usage"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
"github.com/router-for-me/CLIProxyAPI/v7/internal/winutil"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
log "github.com/sirupsen/logrus"
)
+1 -1
View File
@@ -14,7 +14,7 @@ import (
"strings"
"time"
"github.com/router-for-me/CLIProxyAPI/v6/internal/trayicon"
"github.com/router-for-me/CLIProxyAPI/v7/internal/trayicon"
)
func main() {
+137 -43
View File
@@ -10,29 +10,34 @@ import (
"fmt"
"io"
"io/fs"
"net"
"net/url"
"os"
"path/filepath"
"strconv"
"strings"
"time"
"github.com/joho/godotenv"
configaccess "github.com/router-for-me/CLIProxyAPI/v6/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v6/internal/buildinfo"
"github.com/router-for-me/CLIProxyAPI/v6/internal/cmd"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v6/internal/desktopctl"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v6/internal/managementasset"
"github.com/router-for-me/CLIProxyAPI/v6/internal/misc"
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v6/internal/store"
_ "github.com/router-for-me/CLIProxyAPI/v6/internal/translator"
"github.com/router-for-me/CLIProxyAPI/v6/internal/tui"
"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
configaccess "github.com/router-for-me/CLIProxyAPI/v7/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v7/internal/buildinfo"
"github.com/router-for-me/CLIProxyAPI/v7/internal/cmd"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/desktopctl"
"github.com/router-for-me/CLIProxyAPI/v7/internal/home"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v7/internal/managementasset"
"github.com/router-for-me/CLIProxyAPI/v7/internal/misc"
"github.com/router-for-me/CLIProxyAPI/v7/internal/redisqueue"
"github.com/router-for-me/CLIProxyAPI/v7/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v7/internal/startupconfig"
"github.com/router-for-me/CLIProxyAPI/v7/internal/store"
_ "github.com/router-for-me/CLIProxyAPI/v7/internal/translator"
"github.com/router-for-me/CLIProxyAPI/v7/internal/tui"
"github.com/router-for-me/CLIProxyAPI/v7/internal/usage"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
log "github.com/sirupsen/logrus"
)
@@ -105,6 +110,7 @@ func main() {
var switchMode string
var projectID string
var vertexImport string
var vertexImportPrefix string
var configPath string
var password string
var noIncognito bool
@@ -139,6 +145,8 @@ func main() {
// TUI flag
var launchTUI bool
var homeAddr string
var homePassword string
var standalone bool
var localModel bool
@@ -155,8 +163,8 @@ func main() {
flag.BoolVar(&noIncognito, "no-incognito", false, "Force disable incognito mode (uses existing browser session)")
flag.IntVar(&oauthCallbackPort, "oauth-callback-port", 0, "Override OAuth callback port (defaults to provider-specific port)")
flag.BoolVar(&antigravityLogin, "antigravity-login", false, "Login to Antigravity using OAuth")
flag.BoolVar(&kiroLogin, "kiro-login", false, "Login to Kiro using Google OAuth")
flag.BoolVar(&kiroGoogleLogin, "kiro-google-login", false, "Login to Kiro using Google OAuth (same as --kiro-login)")
flag.BoolVar(&kiroLogin, "kiro-login", false, "Login to Kiro using AWS Builder ID")
flag.BoolVar(&kiroGoogleLogin, "kiro-google-login", false, "Show Kiro Google OAuth limitation and supported alternatives")
flag.BoolVar(&kiroAWSLogin, "kiro-aws-login", false, "Login to Kiro using AWS Builder ID (device code flow)")
flag.BoolVar(&kiroAWSAuthCode, "kiro-aws-authcode", false, "Login to Kiro using AWS Builder ID (authorization code flow, better UX)")
flag.BoolVar(&kiroImport, "kiro-import", false, "Import Kiro token from Kiro IDE (~/.aws/sso/cache/kiro-auth-token.json)")
@@ -180,8 +188,11 @@ func main() {
flag.StringVar(&projectID, "project_id", "", "Project ID (Gemini only, not required)")
flag.StringVar(&configPath, "config", DefaultConfigPath, "Configure File Path")
flag.StringVar(&vertexImport, "vertex-import", "", "Import Vertex service account key JSON file")
flag.StringVar(&vertexImportPrefix, "vertex-import-prefix", "", "Prefix for Vertex model namespacing (use with -vertex-import)")
flag.StringVar(&password, "password", "", "")
flag.BoolVar(&launchTUI, "tui", false, "Start with terminal management UI")
flag.StringVar(&homeAddr, "home", "", "Home control plane address in host:port format (loads config from home and skips local config file)")
flag.StringVar(&homePassword, "home-password", "", "Home control plane password (Redis AUTH)")
flag.BoolVar(&standalone, "standalone", false, "In TUI mode, start an embedded local server")
flag.BoolVar(&localModel, "local-model", false, "Use embedded model catalog only, skip remote model fetching")
@@ -299,6 +310,7 @@ func main() {
var err error
var cfg *config.Config
var isCloudDeploy bool
var configLoadedFromHome bool
var (
usePostgresStore bool
pgStoreDSN string
@@ -309,6 +321,7 @@ func main() {
gitStoreRemoteURL string
gitStoreUser string
gitStorePassword string
gitStoreBranch string
gitStoreLocalPath string
gitStoreInst *store.GitTokenStore
gitStoreRoot string
@@ -378,6 +391,9 @@ func main() {
if value, ok := lookupEnv("GITSTORE_LOCAL_PATH", "gitstore_local_path"); ok {
gitStoreLocalPath = value
}
if value, ok := lookupEnv("GITSTORE_GIT_BRANCH", "gitstore_git_branch"); ok {
gitStoreBranch = value
}
if value, ok := lookupEnv("OBJECTSTORE_ENDPOINT", "objectstore_endpoint"); ok {
useObjectStore = true
objectStoreEndpoint = value
@@ -405,7 +421,68 @@ func main() {
// Determine and load the configuration file.
// Prefer the Postgres store when configured, otherwise fallback to git or local files.
var configFilePath string
if usePostgresStore {
if strings.TrimSpace(homeAddr) != "" {
configLoadedFromHome = true
trimmedHomePassword := strings.TrimSpace(homePassword)
host, portStr, errSplit := net.SplitHostPort(strings.TrimSpace(homeAddr))
if errSplit != nil {
log.Errorf("invalid -home address %q (expected host:port): %v", homeAddr, errSplit)
return
}
host = strings.TrimSpace(host)
if host == "" {
log.Errorf("invalid -home address %q: host is empty", homeAddr)
return
}
port, errPort := strconv.Atoi(strings.TrimSpace(portStr))
if errPort != nil || port <= 0 {
log.Errorf("invalid -home address %q: invalid port %q", homeAddr, portStr)
return
}
homeCfg := config.HomeConfig{
Enabled: true,
Host: host,
Port: port,
Password: trimmedHomePassword,
}
homeClient := home.New(homeCfg)
defer homeClient.Close()
ctxHome, cancelHome := context.WithTimeout(context.Background(), 30*time.Second)
raw, errGetConfig := homeClient.GetConfig(ctxHome)
cancelHome()
if errGetConfig != nil {
log.Errorf("failed to fetch config from home: %v", errGetConfig)
return
}
parsed, errParseConfig := config.ParseConfigBytes(raw)
if errParseConfig != nil {
log.Errorf("failed to parse config payload from home: %v", errParseConfig)
return
}
if parsed == nil {
parsed = &config.Config{}
}
parsed.Home = homeCfg
parsed.Port = 8317 // Default to 8317 for home mode, can be overridden by home config
parsed.UsageStatisticsEnabled = true
cfg = parsed
// Keep a non-empty config path for downstream components (log paths, management assets, etc),
// but do not require the file to exist when loading config from home.
if strings.TrimSpace(configPath) != "" {
configFilePath = configPath
} else {
configFilePath = filepath.Join(wd, "config.yaml")
}
// Local stores are intentionally disabled when config is loaded from home.
usePostgresStore = false
useObjectStore = false
useGitStore = false
} else if usePostgresStore {
if pgStoreLocalPath == "" {
pgStoreLocalPath = wd
}
@@ -512,7 +589,7 @@ func main() {
}
gitStoreRoot = filepath.Join(gitStoreLocalPath, "gitstore")
authDir := filepath.Join(gitStoreRoot, "auths")
gitStoreInst = store.NewGitTokenStore(gitStoreRemoteURL, gitStoreUser, gitStorePassword)
gitStoreInst = store.NewGitTokenStore(gitStoreRemoteURL, gitStoreUser, gitStorePassword, gitStoreBranch)
gitStoreInst.SetBaseDir(authDir)
if errRepo := gitStoreInst.EnsureRepository(); errRepo != nil {
log.Errorf("failed to prepare git token store: %v", errRepo)
@@ -550,12 +627,18 @@ func main() {
configFilePath = configPath
cfg, err = config.LoadConfigOptional(configPath, isCloudDeploy)
} else {
wd, err = os.Getwd()
if err != nil {
log.Errorf("failed to get working directory: %v", err)
return
exePath, errExe := os.Executable()
if errExe != nil {
log.WithError(errExe).Debug("failed to resolve executable path for default config lookup")
}
configFilePath = filepath.Join(wd, "config.yaml")
resolution := startupconfig.ResolveConfigPath("", wd, exePath)
if created, errPrepare := startupconfig.EnsureDefaultConfig(resolution); errPrepare != nil {
log.Errorf("failed to prepare default config: %v", errPrepare)
return
} else if created {
log.Infof("default config initialized from template: %s", resolution.ConfigPath)
}
configFilePath = resolution.ConfigPath
cfg, err = config.LoadConfigOptional(configFilePath, isCloudDeploy)
}
if err != nil {
@@ -574,24 +657,27 @@ func main() {
// In cloud deploy mode, check if we have a valid configuration
var configFileExists bool
if isCloudDeploy {
if info, errStat := os.Stat(configFilePath); errStat != nil {
// Don't mislead: API server will not start until configuration is provided.
log.Info("Cloud deploy mode: No configuration file detected; standing by for configuration")
configFileExists = false
} else if info.IsDir() {
log.Info("Cloud deploy mode: Config path is a directory; standing by for configuration")
configFileExists = false
} else if cfg.Port == 0 {
// LoadConfigOptional returns empty config when file is empty or invalid.
// Config file exists but is empty or invalid; treat as missing config
log.Info("Cloud deploy mode: Configuration file is empty or invalid; standing by for valid configuration")
configFileExists = false
if configLoadedFromHome && cfg != nil {
configFileExists = cfg.Port != 0
} else {
log.Info("Cloud deploy mode: Configuration file detected; starting service")
configFileExists = true
if info, errStat := os.Stat(configFilePath); errStat != nil {
// Don't mislead: API server will not start until configuration is provided.
log.Info("Cloud deploy mode: No configuration file detected; standing by for configuration")
configFileExists = false
} else if info.IsDir() {
log.Info("Cloud deploy mode: Config path is a directory; standing by for configuration")
configFileExists = false
} else if cfg.Port == 0 {
// LoadConfigOptional returns empty config when file is empty or invalid.
// Config file exists but is empty or invalid; treat as missing config
log.Info("Cloud deploy mode: Configuration file is empty or invalid; standing by for valid configuration")
configFileExists = false
} else {
log.Info("Cloud deploy mode: Configuration file detected; starting service")
configFileExists = true
}
}
}
// Perform basic semantic validation of the loaded configuration.
if warnings, errValidate := config.ValidateConfig(cfg); errValidate != nil {
log.Errorf("invalid configuration: %v", errValidate)
@@ -603,6 +689,8 @@ func main() {
}
usage.SetStatisticsEnabled(cfg.UsageStatisticsEnabled)
redisqueue.SetUsageStatisticsEnabled(cfg.UsageStatisticsEnabled)
redisqueue.SetRetentionSeconds(cfg.RedisUsageQueueRetentionSeconds)
coreauth.SetQuotaCooldownDisabled(cfg.DisableCooling)
// AntigravityPrimaryEmail removed - field does not exist
@@ -655,7 +743,7 @@ func main() {
if vertexImport != "" {
// Handle Vertex service account import
cmd.DoVertexImport(cfg, vertexImport)
cmd.DoVertexImport(cfg, vertexImport, vertexImportPrefix)
} else if showVersion {
// Version already printed at startup, just exit
return
@@ -839,8 +927,11 @@ func main() {
if standalone {
// Standalone mode: start an embedded local server and connect TUI client to it.
managementasset.StartAutoUpdater(context.Background(), configFilePath)
if !localModel {
misc.StartAntigravityVersionUpdater(context.Background())
if !localModel && !cfg.Home.Enabled {
registry.StartModelsUpdater(context.Background())
} else if cfg.Home.Enabled {
log.Info("Home mode: remote model updates disabled")
}
origStdout := os.Stdout
origStderr := os.Stderr
@@ -908,8 +999,11 @@ func main() {
}
} else {
managementasset.StartAutoUpdater(context.Background(), configFilePath)
if !localModel {
misc.StartAntigravityVersionUpdater(context.Background())
if !localModel && !cfg.Home.Enabled {
registry.StartModelsUpdater(context.Background())
} else if cfg.Home.Enabled {
log.Info("Home mode: remote model updates disabled")
}
if autoGenPassword {
cmd.StartServiceStandalone(cfg, configFilePath, password)
+5 -5
View File
@@ -14,11 +14,11 @@ import (
"golang.org/x/sys/windows/svc/eventlog"
"golang.org/x/sys/windows/svc/mgr"
configaccess "github.com/router-for-me/CLIProxyAPI/v6/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy"
configaccess "github.com/router-for-me/CLIProxyAPI/v7/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy"
)
const serviceName = "ProxyPilot"
+55 -2
View File
@@ -11,6 +11,13 @@ tls:
cert: ""
key: ""
# Optional "home" control plane integration over Redis protocol.
home:
enabled: false
host: "127.0.0.1"
port: 6379
password: ""
# Management API settings
remote-management:
# Whether to allow remote (non-localhost) management access.
@@ -66,6 +73,11 @@ error-logs-max-files: 10
# When false, disable in-memory usage statistics aggregation
usage-statistics-enabled: false
# How long (in seconds) Redis usage queue items are retained in memory for the RESP interface (LPOP/RPOP).
# Note: the in-process Redis RESP usage output is disabled when home.enabled is true.
# Default: 60. Max: 3600.
redis-usage-queue-retention-seconds: 60
# Proxy URL. Supports socks5/http/https protocols. Example: socks5://user:pass@192.168.1.1:1080/
# Per-entry proxy-url also supports "direct" or "none" to bypass both the global proxy-url and environment proxies explicitly.
proxy-url: ""
@@ -87,30 +99,65 @@ max-retry-credentials: 0
# Maximum wait time in seconds for a cooled-down credential before triggering a retry.
max-retry-interval: 30
# When true, disable auth/model cooldown scheduling globally (prevents blackout windows after failure states).
disable-cooling: false
# disable-image-generation supports: false (default), true, or "chat".
# - true: disable image_generation everywhere (also returns 404 for /v1/images/generations and /v1/images/edits).
# - "chat": disable image_generation injection on non-images endpoints, but keep /v1/images/generations and /v1/images/edits enabled.
disable-image-generation: false
# Core auth auto-refresh worker pool size (OAuth/file-based auth token refresh).
# When > 0, overrides the default worker count (16).
# auth-auto-refresh-workers: 16
# Quota exceeded behavior
quota-exceeded:
switch-project: true # Whether to automatically switch to another project when a quota is exceeded
switch-preview-model: true # Whether to automatically switch to a preview model when a quota is exceeded
antigravity-credits: true # Whether to use credits as last-resort fallback when all free-tier auths are exhausted for Claude models
# Routing strategy for selecting credentials when multiple match.
routing:
strategy: "round-robin" # round-robin (default), fill-first
# Enable universal session-sticky routing for all clients.
# Session IDs are extracted from: metadata.user_id (Claude Code session format),
# X-Session-ID, Session_id (Codex), X-Amp-Thread-Id (Amp CLI),
# X-Client-Request-Id (PI), conversation_id, or first few messages hash.
# Automatic failover is always enabled when bound auth becomes unavailable.
session-affinity: false # default: false
# How long session-to-auth bindings are retained. Default: 1h
session-affinity-ttl: "1h"
# When true, enable authentication for the WebSocket API (/v1/ws).
ws-auth: false
# When true, enable Gemini CLI internal endpoints (/v1internal:*).
# Default is false for safety.
enable-gemini-cli-endpoint: false
# When > 0, emit blank lines every N seconds for non-streaming responses to prevent idle timeouts.
nonstream-keepalive-interval: 0
# Streaming behavior (SSE keep-alives + safe bootstrap retries).
# streaming:
# keepalive-seconds: 15 # Default: 0 (disabled). <= 0 disables keep-alives.
# bootstrap-retries: 1 # Default: 0 (disabled). Retries before first byte is sent.
# Signature cache validation for thinking blocks (Antigravity/Claude).
# When true (default), cached signatures are preferred and validated.
# When false, client signatures are used directly after normalization (bypass mode for testing).
# antigravity-signature-cache-enabled: true
# Bypass mode signature validation strictness (only applies when signature cache is disabled).
# When true, validates full Claude protobuf tree (Field 2 -> Field 1 structure).
# When false (default), only checks R/E prefix + base64 + first byte 0x12.
# antigravity-signature-bypass-strict: false
# Gemini API keys
# gemini-api-key:
# - api-key: "AIzaSy...01"
# prefix: "test" # optional: require calls like "test/gemini-3-pro-preview" to target this credential
# disable-cooling: false # optional: per-auth override for auth/model cooldown scheduling
# base-url: "https://generativelanguage.googleapis.com"
# headers:
# X-Custom-Header: "custom-value"
@@ -130,6 +177,7 @@ nonstream-keepalive-interval: 0
# codex-api-key:
# - api-key: "sk-atSM..."
# prefix: "test" # optional: require calls like "test/gpt-5-codex" to target this credential
# disable-cooling: false # optional: per-auth override for auth/model cooldown scheduling
# base-url: "https://www.example.com" # use the custom codex API endpoint
# headers:
# X-Custom-Header: "custom-value"
@@ -149,6 +197,7 @@ nonstream-keepalive-interval: 0
# - api-key: "sk-atSM..." # use the official claude API key, no need to set the base url
# - api-key: "sk-atSM..."
# prefix: "test" # optional: require calls like "test/claude-sonnet-latest" to target this credential
# disable-cooling: false # optional: per-auth override for auth/model cooldown scheduling
# base-url: "https://www.example.com" # use the custom claude API endpoint
# headers:
# X-Custom-Header: "custom-value"
@@ -172,6 +221,8 @@ nonstream-keepalive-interval: 0
# - "API"
# - "proxy"
# cache-user-id: true # optional: default is false; set true to reuse cached user_id per API key instead of generating a random one each request
# experimental-cch-signing: false # optional: default is false; when true, sign the final /v1/messages body using the current Claude Code cch algorithm
# # keep this disabled unless you explicitly need the behavior, so upstream seed changes fall back to legacy proxy behavior
# Default headers for Claude API requests. Update when Claude Code releases new versions.
# In legacy mode, user-agent/package-version/runtime-version/timeout are used as fallbacks
@@ -199,8 +250,10 @@ nonstream-keepalive-interval: 0
# OpenAI compatibility providers
# openai-compatibility:
# - name: "openrouter" # The name of the provider; it will be used in the user agent and other places.
# disabled: false # optional: set to true to disable this provider without removing it
# prefix: "test" # optional: require calls like "test/kimi-k2" to target this provider's credentials
# base-url: "https://openrouter.ai/api/v1" # The base URL of the provider.
# disable-cooling: false # optional: per-provider override for auth/model cooldown scheduling
# headers:
# X-Custom-Header: "custom-value"
# api-key-entries:
@@ -218,7 +271,7 @@ nonstream-keepalive-interval: 0
# # Requests to that alias will round-robin across the upstream names below,
# # and if the chosen upstream fails before producing output, the request will
# # continue with the next upstream model in the same alias pool.
# - name: "qwen3.5-plus"
# - name: "deepseek-v3.1"
# alias: "claude-opus-4.66"
# - name: "glm-5"
# alias: "claude-opus-4.66"
+4 -4
View File
@@ -17,10 +17,10 @@ services:
ports:
- "8317:8317"
- "8085:8085"
- "1455:1455"
- "54545:54545"
- "51121:51121"
- "11451:11451"
- "127.0.0.1:1455:1455"
- "127.0.0.1:54545:54545"
- "127.0.0.1:51121:51121"
- "127.0.0.1:11451:11451"
volumes:
- ${CLI_PROXY_CONFIG_PATH:-./config.yaml}:/ProxyPilot/config.yaml
- ${CLI_PROXY_AUTH_PATH:-./auths}:/root/.cli-proxy-api
+19
View File
@@ -32,6 +32,25 @@ Important:
- Avoid Claude Code “Auth conflict”: do **not** set both `ANTHROPIC_AUTH_TOKEN` and `ANTHROPIC_API_KEY` at the same time.
- Restart Claude Code after editing settings so the new env vars take effect.
## Account risk warning
Claude/Anthropic OAuth, Google/Gemini OAuth, Gemini CLI OAuth, and Antigravity OAuth are unofficial compatibility paths. Providers may treat proxy, automation, shared-token, or third-party OAuth traffic as abnormal usage or a policy violation. These flows can risk quota loss, account restriction, or account suspension.
Do not use production, billing-critical, school/work, or irreplaceable personal accounts with these OAuth paths unless you accept that risk. Prefer official API keys or officially supported developer access for important accounts.
## Claude OAuth quota limitation
Anthropic currently treats third-party app traffic differently from regular Claude.ai or Claude Code subscription usage. If a Claude OAuth request fails with:
```text
Third-party apps now draw from your extra usage, not your plan limits.
Add more at claude.ai/settings/usage and keep going.
```
that response is coming from Anthropic, not ProxyPilot. It means the OAuth credential is valid, but Anthropic is requiring paid extra usage for this third-party API path. ProxyPilot cannot make those requests draw from normal Claude.ai subscription quota.
For Claude through ProxyPilot, either add extra usage in Anthropic settings, use a supported API-key billing path, or route the client to another provider/model alias such as Antigravity, Codex, Gemini, or Kiro.
## Server-side model aliasing (Claude handler)
CLIProxyAPI rewrites common Claude Code model aliases to locally-available models before routing:
+8 -4
View File
@@ -57,12 +57,16 @@ proxypilot --codex-login # Login to Codex using OAuth
proxypilot --claude-login # Login to Claude using OAuth
proxypilot --qwen-login # Login to Qwen using OAuth
proxypilot --antigravity-login # Login to Antigravity using OAuth
proxypilot --kiro-login # Login to Kiro using Google OAuth
proxypilot --kiro-aws-login # Login to Kiro using AWS Builder ID
proxypilot --kiro-login # Login to Kiro using AWS Builder ID
proxypilot --kiro-aws-login # Login to Kiro using AWS Builder ID device code
proxypilot --minimax-login # Add MiniMax API key
proxypilot --zhipu-login # Add Zhipu AI API key
```
Account risk warning: `--claude-login`, `--login` / Gemini, Gemini CLI OAuth, and `--antigravity-login` are unofficial compatibility login paths. Providers may treat proxy, automation, shared-token, or third-party OAuth traffic as abnormal usage or a policy violation. These flows can risk quota loss, account restriction, or account suspension. Prefer official API-key or officially supported developer access for accounts you cannot afford to lose.
Kiro Google/GitHub OAuth is not currently supported because AWS Cognito rejects third-party localhost callbacks. Use AWS Builder ID or import a token from the official Kiro IDE instead.
### Login Options
```bash
@@ -309,8 +313,8 @@ proxypilot --codex-login # Codex OAuth
proxypilot --claude-login # Claude OAuth
proxypilot --qwen-login # Qwen OAuth
proxypilot --antigravity-login # Antigravity OAuth
proxypilot --kiro-login # Kiro Google OAuth
proxypilot --kiro-aws-login # Kiro AWS Builder ID
proxypilot --kiro-login # Kiro AWS Builder ID
proxypilot --kiro-aws-login # Kiro AWS Builder ID device code
# Agent Detection & Setup
proxypilot --detect-agents # Detect installed agents
+2
View File
@@ -61,6 +61,8 @@ ProxyPilot tray app includes:
- One-click login launchers: Antigravity, Gemini CLI, Codex, Claude, Qwen
- `Open Auth Folder` shortcut
Account risk warning: Antigravity, Gemini/Gemini CLI, and Claude login launchers use unofficial compatibility OAuth paths. Providers may treat proxy, automation, shared-token, or third-party OAuth traffic as abnormal usage or a policy violation. These flows can risk quota loss, account restriction, or account suspension. Use disposable or non-critical accounts only if you accept that risk; prefer official API keys or officially supported developer access for important accounts.
## Logs / diagnostics
Useful paths:
+100
View File
@@ -0,0 +1,100 @@
# ProxyPilot Rust Replatform
This branch carries the Rust rewrite line for ProxyPilot.
## Branch contract
- Go `main` remains the shipping line while Rust grows.
- Rust lives under [`rust/`](/home/fsos/code-lean/ai-tools/ProxyPilot/rust).
- The Rust line is terminal-first and does not treat the browser dashboard as a required early target.
- The first real milestone is a working Rust executable that can proxy a Codex-compatible request end-to-end.
## Current milestone
`proxypilot-rs` is the first Rust binary in the repo. It currently provides:
- a `run` command that starts a local proxy server
- a `tui` command that opens a minimal terminal operator view
- a TOML config model plus a separate local state file for saved accounts
- explicit Codex-facing routes for `/v1/models`, `/v1/chat/completions`, `/v1/responses`, `/v1/responses/compact`, and Codex direct `/backend-api/codex/responses` aliases
- a local `/v0/runtime/stats` operator surface for runtime-memory counters, auth-health, and last refresh status
- read-only `/v0/management/status`, `/v0/management/config`, and `/v0/management/accounts` endpoints with API key and token material redacted
- CLI account commands for adding, importing, exporting, device-login, refresh, listing, activating, and removing accounts; these manage saved disk state only and do not pretend a running proxy has already adopted the change
- optional `codex.refresh_token_url` config support for redirecting Codex refresh exchanges to a local stub during validation, while preserving the live OpenAI default when unset
- a TUI operator console that shows live runtime stats, runtime-unavailable copy, local accounts, selected-account details, token state labels, terminal actions for activate/refresh/delete, separate selected-account vs active-account refresh handling, and the `r` / `c` reload-clear controls
- runtime credential resolution that prefers the active saved account over the config fallback key
- proactive Codex token refresh when the active saved account is already expired or within the refresh window, plus the existing 401-retry refresh fallback
- end-to-end tests against a mocked upstream server
## Design defaults
- Go is the behavioral reference, not a line-by-line port target.
- The Rust branch is allowed to redesign config and management surfaces.
- The product identity does not change: ProxyPilot still exists to be a local proxy for coding tools.
## Local workflow
From the repo root:
```bash
cd rust
cargo run -p proxypilot-rs -- init
cargo run -p proxypilot-rs -- account add-codex --name primary --api-key sk-...
cargo run -p proxypilot-rs -- account import-codex --file ../auths/codex-example.json --activate
cargo run -p proxypilot-rs -- account export --file proxypilot-rs.accounts.toml
cargo run -p proxypilot-rs -- account import --file proxypilot-rs.accounts.toml
cargo run -p proxypilot-rs -- account import --file proxypilot-rs.accounts.toml --replace
cargo run -p proxypilot-rs -- account login-codex-device --activate
cargo run -p proxypilot-rs -- account refresh-codex
cargo run -p proxypilot-rs -- account remove --name old-account
cargo run -p proxypilot-rs -- run --config proxypilot-rs.toml
```
Run the local Codex-compatible smoke check without touching real provider
accounts:
```bash
cd rust
./scripts/codex-smoke.sh
```
The smoke check starts a local mock upstream, saves a temporary Codex account,
starts the Rust proxy, and verifies `/healthz`, `/v0/runtime/stats`,
`/v1/models`, `/v1/chat/completions`, and `/v1/responses`.
Run the local Claude API-key smoke check:
```bash
cd rust
./scripts/claude-smoke.sh
```
The Claude smoke starts a local mock Anthropic-compatible upstream, starts the
Rust proxy with `providers.active = "claude"`, and verifies Claude API-key
headers on `/v1/models` and `/v1/messages`.
## Current Codex Gaps
The Rust line is not ready to replace the Go Codex surface yet. Current status:
- `/health` and `/healthz` both return Rust health status.
- `/v1/responses/compact` forwards to the active provider upstream.
- `/backend-api/codex/responses` and `/backend-api/codex/responses/compact`
forward through the Codex-compatible `/v1/responses` paths when the active
provider is `codex`; they reject non-Codex active providers instead of
silently routing a Codex direct alias to Claude or another provider.
- `GET /v1/responses` and `GET /backend-api/codex/responses` return explicit
`501 Not Implemented` responses because Rust websocket response proxying is
still deferred.
- Codex websocket streaming
- Codex image generation and image edit handling
- Browser dashboard and broad `/v0/management/**` parity beyond status, config,
and accounts
- Go model registry and model-mapping behavior
Open the TUI with:
```bash
cd rust
cargo run -p proxypilot-rs -- tui --config proxypilot-rs.toml
```
+26
View File
@@ -0,0 +1,26 @@
# Session Progress - 2026-04-01
## Summary
Merged upstream CLIProxyAPI release `v6.9.7` into ProxyPilot on a dedicated sync branch, carrying forward ProxyPilot branding and local-manual release workflow choices while absorbing the upstream changes from `v6.9.6` and `v6.9.7`.
Re-established a fully local Windows installer build path on Linux using a repo-scoped Wine prefix plus Inno Setup so the `Setup.exe` can be built without CI/CD.
## Upstream Releases Included
- `v6.9.6` - intermediate upstream changes now included through the `v6.9.7` merge
- `v6.9.7` - Codex executor/cache behavior updates, Amp response rewriting improvements, scheduler/auth updates, and related tests
## Conflict Resolution Notes
- Kept the ProxyPilot README and removed upstream `README_CN.md` and `README_JA.md` revivals because this fork intentionally does not ship those translation files.
- Followed upstream on the Codex continuity revert by deleting `internal/runtime/executor/codex_continuity.go` and aligning the Codex cache tests with `v6.9.7`, since upstream explicitly reverted that helper rather than evolving it.
- Preserved the existing fork policy of local/manual releases instead of changing CI or workflow behavior as part of this sync.
## Repo State Updates
- Updated `UPSTREAM_VERSION` to `v6.9.7`.
- Updated `features.json` upstream marker to `CLIProxyAPI v6.9.7`.
- Prepared a local Windows installer toolchain using Wine + Inno Setup for the upcoming `v0.3.0-dev-0.47` release build.
## Pending Verification
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
- full local release asset build for `v0.3.0-dev-0.47`
- `gh release list -R Finesssee/ProxyPilot --limit 5`
+28
View File
@@ -0,0 +1,28 @@
# Session Progress - 2026-04-03
## Summary
Merged upstream CLIProxyAPI release `v6.9.10` into ProxyPilot on a dedicated sync branch and preserved the fork's local/manual release workflow.
Reconciled upstream's executor helper-package move with ProxyPilot's existing executor call sites by keeping the new `internal/runtime/executor/helps/` layout while restoring compatibility entrypoints in `internal/runtime/executor/`.
## Upstream Releases Included
- `v6.9.8` - intermediate upstream changes now included through the `v6.9.10` merge
- `v6.9.9` - intermediate upstream changes now included through the `v6.9.10` merge
- `v6.9.10` - latest upstream sync target for this release cycle
## Conflict Resolution Notes
- Kept the ProxyPilot README and continued to omit upstream `README_CN.md` and `README_JA.md` revivals because this fork does not ship those translation files.
- Preserved ProxyPilot's auth-file persistence behavior, including `prefix`, `proxy_url`, `priority`, and last-refresh handling, while also bringing in upstream custom-header hydration and patch support.
- Kept upstream's `internal/runtime/executor/helps/` reorganization, but restored executor-package compatibility wrappers/files so existing fork-specific executors still compile without a larger refactor in the same release pass.
- Continued to omit the upstream TUI translation/usage-tab files because ProxyPilot's current TUI surface does not carry those forked files.
## Repo State Updates
- Updated `UPSTREAM_VERSION` to `v6.9.10`.
- Updated `features.json` upstream marker to `CLIProxyAPI v6.9.10`.
- Recorded the merge commit `chore: merge upstream CLIProxyAPI v6.9.10` before starting validation and release assembly.
## Pending Verification
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
- full local release asset build for `v0.3.0-dev-0.48`
- `gh release list -R Finesssee/ProxyPilot --limit 5`
+26
View File
@@ -0,0 +1,26 @@
# Session Progress - 2026-04-04
## Summary
Merged upstream CLIProxyAPI release `v6.9.12` into ProxyPilot on a dedicated sync branch and kept the existing local/manual release workflow unchanged.
This catch-up pull included the upstream delta from `v6.9.11` through `v6.9.12` and merged cleanly without any fork-specific conflict resolution work.
## Upstream Releases Included
- `v6.9.11` - intermediate upstream changes now included through the `v6.9.12` merge
- `v6.9.12` - latest upstream sync target for this release cycle
## Merge Notes
- The merge applied cleanly with upstream changes concentrated in `config.example.yaml`, `internal/config/sdk_config.go`, `internal/runtime/executor/qwen_executor.go`, and `sdk/api/handlers/gemini/gemini-cli_handlers.go`.
- No additional ProxyPilot compatibility shims were needed beyond the executor-wrapper structure restored in the prior sync.
- ProxyPilot branding, release routing, and local-build release flow remain unchanged.
## Repo State Updates
- Updated `UPSTREAM_VERSION` to `v6.9.12`.
- Updated `features.json` upstream marker to `CLIProxyAPI v6.9.12`.
- Prepared the repo for the next local/manual release: `v0.3.0-dev-0.49`.
## Pending Verification
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
- full local release asset build for `v0.3.0-dev-0.49`
- `gh release list -R Finesssee/ProxyPilot --limit 5`
+28
View File
@@ -0,0 +1,28 @@
# Session Progress - 2026-04-05
## Summary
Merged upstream CLIProxyAPI release `v6.9.15` into ProxyPilot on a dedicated sync branch based on the latest `origin/main`.
Kept the local management-asset security fix from `origin/main`, preserved ProxyPilot branding and local/manual release workflow choices, and absorbed the upstream delta from `v6.9.13` through `v6.9.15`.
## Upstream Releases Included
- `v6.9.13` - vertex credential namespacing support, websocket custom tool-call repair, and model catalog refreshes
- `v6.9.14` - antigravity version refresh updates and additional upstream maintenance changes
- `v6.9.15` - API-key proxy resolution improvements and stricter Qwen system-message handling
## Conflict Resolution Notes
- Kept the ProxyPilot README branding and continued to omit upstream `README_CN.md` and `README_JA.md` revivals because this fork does not ship those translated files.
- Kept the existing local/manual release workflow and combined the fork's broader `.gitignore` with upstream's additional local-tool entries.
- Preserved the local `internal/managementasset/updater.go` security behavior that removes the unverified fallback download path while accepting the rest of the upstream sync.
- Merged `cmd/server/main.go` semantically so ProxyPilot keeps its wider CLI surface while also adding upstream `--vertex-import-prefix` support.
## Repo State Updates
- Updated `UPSTREAM_VERSION` to `v6.9.15`.
- Updated `features.json` upstream marker to `CLIProxyAPI v6.9.15`.
- Prepared the repo for the next clean release tag: `v0.3.0-dev-0.51`.
## Pending Verification
- `go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `go vet ./...`
- `go test ./...`
- clean-tree packaging and release asset assembly for `v0.3.0-dev-0.51`
- `gh release list -R Finesssee/ProxyPilot --limit 5`
+9
View File
@@ -0,0 +1,9 @@
# Session 2026-04-11
- Targeted upstream sync: `CLIProxyAPI v6.9.22`
- Merge base from shipping line: `v0.3.0-dev-0.53` on `main`
- Upstream merge branch: `sync-upstream-v6.9.18-to-v6.9.22`
- Conflict resolution:
- kept ProxyPilot-local `.gitignore` entries while accepting upstream code changes
- updated upstream tracking markers and release metadata to `v6.9.22`
- Planned release target after validation: `v0.3.0-dev-0.54`
+13
View File
@@ -0,0 +1,13 @@
# Session 2026-04-12
- Target upstream release: `CLIProxyAPI v6.9.23` (`0ab1f54173d477d6a39457b6aaf1f663b315fdb2`)
- Base local release line before merge: `ProxyPilot v0.3.0-dev-0.54` on `main`
- Merge branch: `sync-upstream-v6.9.22-to-v6.9.23`
- Conflict resolution:
- kept ProxyPilot branding and docs flow in `README.md`
- kept upstream Qwen executor/runtime changes from `v6.9.23`
- did not revive upstream locale README files (`README_CN.md`, `README_JA.md`)
- Validation plan:
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
+12
View File
@@ -0,0 +1,12 @@
# Session 2026-04-13
- Target upstream release: `CLIProxyAPI v6.9.24` (`c4459c43462c45bcd8ac346c72b98647675a014e`)
- Base local release line before merge: `ProxyPilot main` at `v6.9.23`
- Merge branch: `sync-upstream-v6.9.23-to-v6.9.24`
- Merge result:
- clean merge with no manual conflict resolution required
- kept ProxyPilot fork structure intact while taking upstream auth auto-refresh loop and Antigravity fixes
- Validation plan:
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
+14
View File
@@ -0,0 +1,14 @@
# Session 2026-04-16
- Target upstream release: `CLIProxyAPI v6.9.27` (`ba454dbfbfb41c3a8218118323f828ea19e92c3d`)
- Base local release line before merge: `ProxyPilot main` at `v6.9.24` (`06c3e954`)
- Merge branch: `sync-upstream-v6.9.24-to-v6.9.27`
- Conflict resolution:
- kept ProxyPilot branding and docs flow in `README.md`
- did not revive upstream locale README files (`README_CN.md`, `README_JA.md`)
- preserved ProxyPilot Qwen auth/runtime support while taking upstream session-affinity and Antigravity fixes
- kept local request execution metadata behavior in `sdk/api/handlers/handlers.go` so pinned-auth selection remains stable
- Validation:
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
+14
View File
@@ -0,0 +1,14 @@
# Session 2026-04-17
- Target upstream release: `CLIProxyAPI v6.9.28` (`5dcca69ea5be9de76e04c7f5dcd51098e7fc9018`)
- Base local release line before merge: `ProxyPilot main` at `v6.9.27` (`71be5e89`)
- Merge branch: `sync-upstream-v6.9.27-to-v6.9.28`
- Conflict resolution:
- kept ProxyPilot branding and docs flow in `README.md`
- did not revive upstream locale README files (`README_CN.md`, `README_JA.md`)
- preserved ProxyPilot `qwen` and `iflow` auth/runtime/config/model support while taking upstream auth proxy-override support and Claude Opus 4.7 model catalog updates
- kept local synthetic idempotency-key behavior in `sdk/api/handlers/handlers.go` and updated the new upstream metadata test to match fork behavior
- Validation:
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
+14
View File
@@ -0,0 +1,14 @@
# Session 2026-04-18
- Target upstream release: `CLIProxyAPI v6.9.29` (`c6baa64b4e3cf85f8400e2cc9d9bd3d7040db187`)
- Base local release line before merge: `ProxyPilot main` at `v6.9.28` (`7ec328f9`)
- Merge branch: `sync-upstream-v6.9.28-to-v6.9.29`
- Conflict resolution:
- preserved ProxyPilot's clone-and-rollback config persistence behavior in `internal/api/handlers/management/handler.go` while integrating upstream `auth-index` management handlers
- took upstream Claude executor test expectations that tighten forwarded system-prompt behavior and strict-mode assertions
- no provider-surface carry work was needed this round; local `qwen`, `iflow`, and synthetic idempotency behavior remained intact
- Validation:
- `go test ./internal/api/handlers/management ./internal/runtime/executor`
- `mise exec go@1.26.0 -- go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `mise exec go@1.26.0 -- go vet ./...`
- `mise exec go@1.26.0 -- go test ./...`
+13
View File
@@ -0,0 +1,13 @@
# Session 2026-04-24
- Target upstream release: `CLIProxyAPI v6.9.36` (`f1ba6151a99240902bcda12102c921b0ead01d2d`)
- Base local release line before merge: `ProxyPilot main` at upstream `v6.9.30` (`2ebbb596`)
- Merge branch: `sync-upstream-v6.9.30-to-v6.9.36`
- Conflict resolution:
- no textual conflicts in the upstream merge; existing ProxyPilot fork surfaces carried forward cleanly
- corrected stale upstream marker files from the prior `v6.9.30` sync so `.upstream-sync-tag` and `.upstream-sync-sha` now match the merged upstream release
- kept the existing fork behavior around local provider support while taking upstream Codex image-generation, Antigravity credits fallback, logging, and `/healthz` improvements
- Validation:
- `go test ./...`
- `go build ./cmd/server ./cmd/proxypilot-tray ./cmd/proxypilotpack`
- `go vet ./...`
+8 -8
View File
@@ -24,14 +24,14 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/api"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
clipexec "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/logging"
sdktr "github.com/router-for-me/CLIProxyAPI/v6/sdk/translator"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/api"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
clipexec "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/executor"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/config"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/logging"
sdktr "github.com/router-for-me/CLIProxyAPI/v7/sdk/translator"
)
const (
+2 -2
View File
@@ -16,8 +16,8 @@ import (
"strings"
"time"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
clipexec "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
clipexec "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/executor"
log "github.com/sirupsen/logrus"
)
+2 -2
View File
@@ -4,8 +4,8 @@ import (
"context"
"fmt"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/translator"
_ "github.com/router-for-me/CLIProxyAPI/v6/sdk/translator/builtin"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/translator"
_ "github.com/router-for-me/CLIProxyAPI/v7/sdk/translator/builtin"
)
func main() {
+1 -1
View File
@@ -1,7 +1,7 @@
{
"project": "ProxyPilot",
"version": "0.3.0",
"upstream": "CLIProxyAPI v6.9.5",
"upstream": "CLIProxyAPI v6.9.36",
"features": [
{
"id": 1,
+9 -3
View File
@@ -1,4 +1,4 @@
module github.com/router-for-me/CLIProxyAPI/v6
module github.com/router-for-me/CLIProxyAPI/v7
go 1.26.0
@@ -22,6 +22,7 @@ require (
github.com/pelletier/go-toml/v2 v2.2.2
github.com/pkg/browser v0.0.0-20240102092130-5ac0b6a4141c
github.com/prometheus/client_golang v1.23.2
github.com/redis/go-redis/v9 v9.19.0
github.com/refraction-networking/utls v1.8.2
github.com/sirupsen/logrus v1.9.3
github.com/stretchr/testify v1.11.1
@@ -38,6 +39,11 @@ require (
gopkg.in/yaml.v3 v3.0.1
)
require (
github.com/cespare/xxhash/v2 v2.3.0 // indirect
go.uber.org/atomic v1.11.0 // indirect
)
require (
cloud.google.com/go/compute/metadata v0.3.0 // indirect
github.com/Microsoft/go-winio v0.6.2 // indirect
@@ -45,7 +51,6 @@ require (
github.com/beorn7/perks v1.0.1 // indirect
github.com/bytedance/sonic v1.11.6 // indirect
github.com/bytedance/sonic/loader v0.1.1 // indirect
github.com/cespare/xxhash/v2 v2.3.0 // indirect
github.com/charmbracelet/colorprofile v0.2.3-0.20250311203215-f60798e515dc // indirect
github.com/charmbracelet/x/ansi v0.10.1 // indirect
github.com/charmbracelet/x/cellbuf v0.0.13-0.20250311204145-2c3ea96c31dd // indirect
@@ -96,6 +101,7 @@ require (
github.com/muesli/termenv v0.16.0 // indirect
github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
github.com/oxtoacart/bpool v0.0.0-20190530202638-03653db5a59c // indirect
github.com/pierrec/xxHash v0.1.5
github.com/pjbgf/sha1cd v0.5.0 // indirect
github.com/pmezard/go-difflib v1.0.0 // indirect
github.com/prometheus/client_model v0.6.2 // indirect
@@ -113,6 +119,6 @@ require (
golang.org/x/arch v0.8.0 // indirect
golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b // indirect
golang.org/x/text v0.31.0 // indirect
google.golang.org/protobuf v1.36.8 // indirect
google.golang.org/protobuf v1.36.8
gopkg.in/ini.v1 v1.67.0 // indirect
)
+12
View File
@@ -16,6 +16,10 @@ github.com/aymanbagabas/go-udiff v0.2.0 h1:TK0fH4MteXUDspT88n8CKzvK0X9O2xu9yQjWp
github.com/aymanbagabas/go-udiff v0.2.0/go.mod h1:RE4Ex0qsGkTAJoQdQQCA0uG+nAzJO/pI/QwceO5fgrA=
github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw=
github.com/bsm/ginkgo/v2 v2.12.0 h1:Ny8MWAHyOepLGlLKYmXG4IEkioBysk6GpaRTLC8zwWs=
github.com/bsm/ginkgo/v2 v2.12.0/go.mod h1:SwYbGRRDovPVboqFv0tPTcG1sN61LM1Z4ARdbAV9g4c=
github.com/bsm/gomega v1.27.10 h1:yeMWxP2pV2fG3FgAODIY8EiRE3dy0aeFYt4l7wh6yKA=
github.com/bsm/gomega v1.27.10/go.mod h1:JyEr/xRbxbtgWNi8tIEVPUYZ5Dzef52k01W3YH0H+O0=
github.com/bytedance/sonic v1.11.6 h1:oUp34TzMlL+OY1OUWxHqsdkgC/Zfc85zGqw9siXjrc0=
github.com/bytedance/sonic v1.11.6/go.mod h1:LysEHSvpvDySVdC2f87zGWf6CIKJcAvqab1ZaiQtds4=
github.com/bytedance/sonic/loader v0.1.1 h1:c+e5Pt1k/cy5wMveRDyk2X4B9hF4g7an8N3zCYjJFNM=
@@ -180,6 +184,8 @@ github.com/oxtoacart/bpool v0.0.0-20190530202638-03653db5a59c h1:rp5dCmg/yLR3mgF
github.com/oxtoacart/bpool v0.0.0-20190530202638-03653db5a59c/go.mod h1:X07ZCGwUbLaax7L0S3Tw4hpejzu63ZrrQiUe6W0hcy0=
github.com/pelletier/go-toml/v2 v2.2.2 h1:aYUidT7k73Pcl9nb2gScu7NSrKCSHIDE89b3+6Wq+LM=
github.com/pelletier/go-toml/v2 v2.2.2/go.mod h1:1t835xjRzz80PqgE6HHgN2JOsmgYu/h4qDAS4n929Rs=
github.com/pierrec/xxHash v0.1.5 h1:n/jBpwTHiER4xYvK3/CdPVnLDPchj8eTJFFLUb4QHBo=
github.com/pierrec/xxHash v0.1.5/go.mod h1:w2waW5Zoa/Wc4Yqe0wgrIYAGKqRMf7czn2HNKXmuL+I=
github.com/pjbgf/sha1cd v0.5.0 h1:a+UkboSi1znleCDUNT3M5YxjOnN1fz2FhN48FlwCxs0=
github.com/pjbgf/sha1cd v0.5.0/go.mod h1:lhpGlyHLpQZoxMv8HcgXvZEhcGs0PG/vsZnEJ7H0iCM=
github.com/pkg/browser v0.0.0-20240102092130-5ac0b6a4141c h1:+mdjkGKdHQG3305AYmdv1U2eRNDiU2ErMBj1gwrq8eQ=
@@ -194,6 +200,8 @@ github.com/prometheus/common v0.66.1 h1:h5E0h5/Y8niHc5DlaLlWLArTQI7tMrsfQjHV+d9Z
github.com/prometheus/common v0.66.1/go.mod h1:gcaUsgf3KfRSwHY4dIMXLPV0K/Wg1oZ8+SbZk/HH/dA=
github.com/prometheus/procfs v0.16.1 h1:hZ15bTNuirocR6u0JZ6BAHHmwS1p8B4P6MRqxtzMyRg=
github.com/prometheus/procfs v0.16.1/go.mod h1:teAbpZRB1iIAJYREa1LsoWUXykVXA1KlTmWl8x/U+Is=
github.com/redis/go-redis/v9 v9.19.0 h1:XPVaaPSnG6RhYf7p+rmSa9zZfeVAnWsH5h3lxthOm/k=
github.com/redis/go-redis/v9 v9.19.0/go.mod h1:v/M13XI1PVCDcm01VtPFOADfZtHf8YW3baQf57KlIkA=
github.com/refraction-networking/utls v1.8.2 h1:j4Q1gJj0xngdeH+Ox/qND11aEfhpgoEvV+S9iJ2IdQo=
github.com/refraction-networking/utls v1.8.2/go.mod h1:jkSOEkLqn+S/jtpEHPOsVv/4V4EVnelwbMQl4vCWXAM=
github.com/rivo/uniseg v0.2.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
@@ -239,6 +247,10 @@ github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65E
github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e h1:JVG44RsyaB9T2KIHavMF/ppJZNG9ZpyihvCd0w101no=
github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e/go.mod h1:RbqR21r5mrJuqunuUZ/Dhy/avygyECGrLceyNeo4LiM=
github.com/zeebo/xxh3 v1.1.0 h1:s7DLGDK45Dyfg7++yxI0khrfwq9661w9EN78eP/UZVs=
github.com/zeebo/xxh3 v1.1.0/go.mod h1:IisAie1LELR4xhVinxWS5+zf1lA4p0MW4T+w+W07F5s=
go.uber.org/atomic v1.11.0 h1:ZvwS0R+56ePWxUNi+Atn9dWONBPp/AUETXlHW0DxSjE=
go.uber.org/atomic v1.11.0/go.mod h1:LUxbIzbOniOlMKjJjyPfpl4v+PKK2cNJn91OQbhoJI0=
go.uber.org/goleak v1.3.0 h1:2K3zAYmnTNqV73imy9J1T3WC+gmCePx2hEGkimedGto=
go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE=
go.yaml.in/yaml/v2 v2.4.2 h1:DzmwEr2rDGHl7lsFgAHxmNz/1NlQ7xLIrlN2h5d1eGI=
+2 -2
View File
@@ -5,8 +5,8 @@ import (
"net/http"
"strings"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v6/sdk/access"
sdkconfig "github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v7/sdk/access"
sdkconfig "github.com/router-for-me/CLIProxyAPI/v7/sdk/config"
)
// Register ensures the config-access provider is available to the access manager.
+3 -3
View File
@@ -6,9 +6,9 @@ import (
"sort"
"strings"
configaccess "github.com/router-for-me/CLIProxyAPI/v6/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v6/sdk/access"
configaccess "github.com/router-for-me/CLIProxyAPI/v7/internal/access/config_access"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v7/sdk/access"
log "github.com/sirupsen/logrus"
)
+32
View File
@@ -0,0 +1,32 @@
package api
import (
"bufio"
"crypto/tls"
"net"
)
type bufferedConn struct {
net.Conn
reader *bufio.Reader
}
func (c *bufferedConn) Read(p []byte) (int, error) {
if c == nil {
return 0, net.ErrClosed
}
if c.reader == nil {
return c.Conn.Read(p)
}
return c.reader.Read(p)
}
func (c *bufferedConn) ConnectionState() tls.ConnectionState {
if c == nil || c.Conn == nil {
return tls.ConnectionState{}
}
if stater, ok := c.Conn.(interface{ ConnectionState() tls.ConnectionState }); ok {
return stater.ConnectionState()
}
return tls.ConnectionState{}
}
+2 -2
View File
@@ -8,8 +8,8 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/api/middleware"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/api/middleware"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
)
// mockCoreManager implements memory.CoreManagerExecutor for testing
@@ -4,7 +4,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/agents"
"github.com/router-for-me/CLIProxyAPI/v7/internal/agents"
)
// AgentsV2Handler provides the new unified agent configuration API
@@ -0,0 +1,107 @@
package management
import (
"net/http"
"strings"
"time"
"github.com/gin-gonic/gin"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
type apiKeyUsageEntry struct {
Success int64 `json:"success"`
Failed int64 `json:"failed"`
RecentRequests []coreauth.RecentRequestBucket `json:"recent_requests"`
}
func mergeRecentRequestBuckets(dst, src []coreauth.RecentRequestBucket) []coreauth.RecentRequestBucket {
if len(dst) == 0 {
return src
}
if len(src) == 0 {
return dst
}
if len(dst) != len(src) {
n := len(dst)
if len(src) < n {
n = len(src)
}
for i := 0; i < n; i++ {
dst[i].Success += src[i].Success
dst[i].Failed += src[i].Failed
}
return dst
}
for i := range dst {
dst[i].Success += src[i].Success
dst[i].Failed += src[i].Failed
}
return dst
}
// GetAPIKeyUsage returns recent request buckets for all in-memory api_key auths,
// grouped by provider and keyed by "base_url|api_key".
func (h *Handler) GetAPIKeyUsage(c *gin.Context) {
if h == nil {
c.JSON(http.StatusInternalServerError, gin.H{"error": "handler not initialized"})
return
}
h.mu.Lock()
manager := h.authManager
h.mu.Unlock()
if manager == nil {
c.JSON(http.StatusServiceUnavailable, gin.H{"error": "core auth manager unavailable"})
return
}
now := time.Now()
out := make(map[string]map[string]apiKeyUsageEntry)
for _, auth := range manager.List() {
if auth == nil {
continue
}
kind, apiKey := auth.AccountInfo()
if !strings.EqualFold(strings.TrimSpace(kind), "api_key") {
continue
}
apiKey = strings.TrimSpace(apiKey)
if apiKey == "" {
continue
}
baseURL := ""
if auth.Attributes != nil {
baseURL = strings.TrimSpace(auth.Attributes["base_url"])
if baseURL == "" {
baseURL = strings.TrimSpace(auth.Attributes["base-url"])
}
}
compositeKey := baseURL + "|" + apiKey
provider := strings.ToLower(strings.TrimSpace(auth.Provider))
if provider == "" {
provider = "unknown"
}
recent := auth.RecentRequestsSnapshot(now)
providerBucket, ok := out[provider]
if !ok {
providerBucket = make(map[string]apiKeyUsageEntry)
out[provider] = providerBucket
}
if existing, exists := providerBucket[compositeKey]; exists {
existing.Success += auth.Success
existing.Failed += auth.Failed
existing.RecentRequests = mergeRecentRequestBuckets(existing.RecentRequests, recent)
providerBucket[compositeKey] = existing
continue
}
providerBucket[compositeKey] = apiKeyUsageEntry{
Success: auth.Success,
Failed: auth.Failed,
RecentRequests: recent,
}
}
c.JSON(http.StatusOK, out)
}
@@ -0,0 +1,95 @@
package management
import (
"context"
"encoding/json"
"net/http"
"net/http/httptest"
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
func sumRecentRequestBuckets(buckets []coreauth.RecentRequestBucket) (int64, int64) {
var success int64
var failed int64
for _, bucket := range buckets {
success += bucket.Success
failed += bucket.Failed
}
return success, failed
}
func TestGetAPIKeyUsage_GroupsByProviderAndAPIKey(t *testing.T) {
t.Setenv("MANAGEMENT_PASSWORD", "")
gin.SetMode(gin.TestMode)
manager := coreauth.NewManager(nil, nil, nil)
if _, err := manager.Register(context.Background(), &coreauth.Auth{
ID: "codex-auth",
Provider: "codex",
Attributes: map[string]string{
"api_key": "codex-key",
"base_url": "https://codex.example.com",
},
}); err != nil {
t.Fatalf("register codex auth: %v", err)
}
if _, err := manager.Register(context.Background(), &coreauth.Auth{
ID: "claude-auth",
Provider: "claude",
Attributes: map[string]string{
"api_key": "claude-key",
"base_url": "https://claude.example.com",
},
}); err != nil {
t.Fatalf("register claude auth: %v", err)
}
manager.MarkResult(context.Background(), coreauth.Result{AuthID: "codex-auth", Provider: "codex", Model: "gpt-5", Success: true})
manager.MarkResult(context.Background(), coreauth.Result{AuthID: "codex-auth", Provider: "codex", Model: "gpt-5", Success: false})
manager.MarkResult(context.Background(), coreauth.Result{AuthID: "claude-auth", Provider: "claude", Model: "claude-4", Success: true})
h := NewHandlerWithoutConfigFilePath(&config.Config{AuthDir: t.TempDir()}, manager)
rec := httptest.NewRecorder()
ginCtx, _ := gin.CreateTestContext(rec)
req := httptest.NewRequest(http.MethodGet, "/v0/management/api-key-usage", nil)
ginCtx.Request = req
h.GetAPIKeyUsage(ginCtx)
if rec.Code != http.StatusOK {
t.Fatalf("status = %d, want %d body=%s", rec.Code, http.StatusOK, rec.Body.String())
}
var payload map[string]map[string]apiKeyUsageEntry
if err := json.Unmarshal(rec.Body.Bytes(), &payload); err != nil {
t.Fatalf("decode payload: %v", err)
}
codexEntry := payload["codex"]["https://codex.example.com|codex-key"]
if codexEntry.Success != 1 || codexEntry.Failed != 1 {
t.Fatalf("codex totals = %d/%d, want 1/1", codexEntry.Success, codexEntry.Failed)
}
if len(codexEntry.RecentRequests) != 20 {
t.Fatalf("codex buckets len = %d, want 20", len(codexEntry.RecentRequests))
}
codexSuccess, codexFailed := sumRecentRequestBuckets(codexEntry.RecentRequests)
if codexSuccess != 1 || codexFailed != 1 {
t.Fatalf("codex totals = %d/%d, want 1/1", codexSuccess, codexFailed)
}
claudeEntry := payload["claude"]["https://claude.example.com|claude-key"]
if claudeEntry.Success != 1 || claudeEntry.Failed != 0 {
t.Fatalf("claude totals = %d/%d, want 1/0", claudeEntry.Success, claudeEntry.Failed)
}
if len(claudeEntry.RecentRequests) != 20 {
t.Fatalf("claude buckets len = %d, want 20", len(claudeEntry.RecentRequests))
}
claudeSuccess, claudeFailed := sumRecentRequestBuckets(claudeEntry.RecentRequests)
if claudeSuccess != 1 || claudeFailed != 0 {
t.Fatalf("claude totals = %d/%d, want 1/0", claudeSuccess, claudeFailed)
}
}
+129 -3
View File
@@ -11,9 +11,10 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/runtime/geminicli"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/proxyutil"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/runtime/geminicli"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/proxyutil"
log "github.com/sirupsen/logrus"
"golang.org/x/oauth2"
"golang.org/x/oauth2/google"
@@ -636,6 +637,11 @@ func (h *Handler) apiCallTransport(auth *coreauth.Auth) http.RoundTripper {
if proxyStr := strings.TrimSpace(auth.ProxyURL); proxyStr != "" {
proxyCandidates = append(proxyCandidates, proxyStr)
}
if h != nil && h.cfg != nil {
if proxyStr := strings.TrimSpace(proxyURLFromAPIKeyConfig(h.cfg, auth)); proxyStr != "" {
proxyCandidates = append(proxyCandidates, proxyStr)
}
}
}
if h != nil && h.cfg != nil {
if proxyStr := strings.TrimSpace(h.cfg.ProxyURL); proxyStr != "" {
@@ -658,6 +664,126 @@ func (h *Handler) apiCallTransport(auth *coreauth.Auth) http.RoundTripper {
return clone
}
type apiKeyConfigEntry interface {
GetAPIKey() string
GetBaseURL() string
}
func resolveAPIKeyConfig[T apiKeyConfigEntry](entries []T, auth *coreauth.Auth) *T {
if auth == nil || len(entries) == 0 {
return nil
}
attrKey, attrBase := "", ""
if auth.Attributes != nil {
attrKey = strings.TrimSpace(auth.Attributes["api_key"])
attrBase = strings.TrimSpace(auth.Attributes["base_url"])
}
for i := range entries {
entry := &entries[i]
cfgKey := strings.TrimSpace((*entry).GetAPIKey())
cfgBase := strings.TrimSpace((*entry).GetBaseURL())
if attrKey != "" && attrBase != "" {
if strings.EqualFold(cfgKey, attrKey) && strings.EqualFold(cfgBase, attrBase) {
return entry
}
continue
}
if attrKey != "" && strings.EqualFold(cfgKey, attrKey) {
if cfgBase == "" || strings.EqualFold(cfgBase, attrBase) {
return entry
}
}
if attrKey == "" && attrBase != "" && strings.EqualFold(cfgBase, attrBase) {
return entry
}
}
if attrKey != "" {
for i := range entries {
entry := &entries[i]
if strings.EqualFold(strings.TrimSpace((*entry).GetAPIKey()), attrKey) {
return entry
}
}
}
return nil
}
func proxyURLFromAPIKeyConfig(cfg *config.Config, auth *coreauth.Auth) string {
if cfg == nil || auth == nil {
return ""
}
authKind, authAccount := auth.AccountInfo()
if !strings.EqualFold(strings.TrimSpace(authKind), "api_key") {
return ""
}
attrs := auth.Attributes
compatName := ""
providerKey := ""
if len(attrs) > 0 {
compatName = strings.TrimSpace(attrs["compat_name"])
providerKey = strings.TrimSpace(attrs["provider_key"])
}
if compatName != "" || strings.EqualFold(strings.TrimSpace(auth.Provider), "openai-compatibility") {
return resolveOpenAICompatAPIKeyProxyURL(cfg, auth, strings.TrimSpace(authAccount), providerKey, compatName)
}
switch strings.ToLower(strings.TrimSpace(auth.Provider)) {
case "gemini":
if entry := resolveAPIKeyConfig(cfg.GeminiKey, auth); entry != nil {
return strings.TrimSpace(entry.ProxyURL)
}
case "claude":
if entry := resolveAPIKeyConfig(cfg.ClaudeKey, auth); entry != nil {
return strings.TrimSpace(entry.ProxyURL)
}
case "codex":
if entry := resolveAPIKeyConfig(cfg.CodexKey, auth); entry != nil {
return strings.TrimSpace(entry.ProxyURL)
}
}
return ""
}
func resolveOpenAICompatAPIKeyProxyURL(cfg *config.Config, auth *coreauth.Auth, apiKey, providerKey, compatName string) string {
if cfg == nil || auth == nil {
return ""
}
apiKey = strings.TrimSpace(apiKey)
if apiKey == "" {
return ""
}
candidates := make([]string, 0, 3)
if v := strings.TrimSpace(compatName); v != "" {
candidates = append(candidates, v)
}
if v := strings.TrimSpace(providerKey); v != "" {
candidates = append(candidates, v)
}
if v := strings.TrimSpace(auth.Provider); v != "" {
candidates = append(candidates, v)
}
for i := range cfg.OpenAICompatibility {
compat := &cfg.OpenAICompatibility[i]
if compat.Disabled {
continue
}
for _, candidate := range candidates {
if candidate != "" && strings.EqualFold(strings.TrimSpace(candidate), compat.Name) {
for j := range compat.APIKeyEntries {
entry := &compat.APIKeyEntries[j]
if strings.EqualFold(strings.TrimSpace(entry.APIKey), apiKey) {
return strings.TrimSpace(entry.ProxyURL)
}
}
return ""
}
}
}
return ""
}
func buildProxyTransport(proxyStr string) *http.Transport {
transport, _, errBuild := proxyutil.BuildHTTPTransport(proxyStr)
if errBuild != nil {
@@ -5,9 +5,9 @@ import (
"net/http"
"testing"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
sdkconfig "github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
sdkconfig "github.com/router-for-me/CLIProxyAPI/v7/sdk/config"
)
func TestAPICallTransportDirectBypassesGlobalProxy(t *testing.T) {
@@ -58,6 +58,105 @@ func TestAPICallTransportInvalidAuthFallsBackToGlobalProxy(t *testing.T) {
}
}
func TestAPICallTransportAPIKeyAuthFallsBackToConfigProxyURL(t *testing.T) {
t.Parallel()
h := &Handler{
cfg: &config.Config{
SDKConfig: sdkconfig.SDKConfig{ProxyURL: "http://global-proxy.example.com:8080"},
GeminiKey: []config.GeminiKey{{
APIKey: "gemini-key",
ProxyURL: "http://gemini-proxy.example.com:8080",
}},
ClaudeKey: []config.ClaudeKey{{
APIKey: "claude-key",
ProxyURL: "http://claude-proxy.example.com:8080",
}},
CodexKey: []config.CodexKey{{
APIKey: "codex-key",
ProxyURL: "http://codex-proxy.example.com:8080",
}},
OpenAICompatibility: []config.OpenAICompatibility{{
Name: "bohe",
BaseURL: "https://bohe.example.com",
APIKeyEntries: []config.OpenAICompatibilityAPIKey{{
APIKey: "compat-key",
ProxyURL: "http://compat-proxy.example.com:8080",
}},
}},
},
}
cases := []struct {
name string
auth *coreauth.Auth
wantProxy string
}{
{
name: "gemini",
auth: &coreauth.Auth{
Provider: "gemini",
Attributes: map[string]string{"api_key": "gemini-key"},
},
wantProxy: "http://gemini-proxy.example.com:8080",
},
{
name: "claude",
auth: &coreauth.Auth{
Provider: "claude",
Attributes: map[string]string{"api_key": "claude-key"},
},
wantProxy: "http://claude-proxy.example.com:8080",
},
{
name: "codex",
auth: &coreauth.Auth{
Provider: "codex",
Attributes: map[string]string{"api_key": "codex-key"},
},
wantProxy: "http://codex-proxy.example.com:8080",
},
{
name: "openai-compatibility",
auth: &coreauth.Auth{
Provider: "bohe",
Attributes: map[string]string{
"api_key": "compat-key",
"compat_name": "bohe",
"provider_key": "bohe",
},
},
wantProxy: "http://compat-proxy.example.com:8080",
},
}
for _, tc := range cases {
tc := tc
t.Run(tc.name, func(t *testing.T) {
t.Parallel()
transport := h.apiCallTransport(tc.auth)
httpTransport, ok := transport.(*http.Transport)
if !ok {
t.Fatalf("transport type = %T, want *http.Transport", transport)
}
req, errRequest := http.NewRequest(http.MethodGet, "https://example.com", nil)
if errRequest != nil {
t.Fatalf("http.NewRequest returned error: %v", errRequest)
}
proxyURL, errProxy := httpTransport.Proxy(req)
if errProxy != nil {
t.Fatalf("httpTransport.Proxy returned error: %v", errProxy)
}
if proxyURL == nil || proxyURL.String() != tc.wantProxy {
t.Fatalf("proxy URL = %v, want %s", proxyURL, tc.wantProxy)
}
})
}
}
func TestAuthByIndexDistinguishesSharedAPIKeysAcrossProviders(t *testing.T) {
t.Parallel()
+123 -47
View File
@@ -22,19 +22,19 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/antigravity"
"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/claude"
"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/codex"
geminiAuth "github.com/router-for-me/CLIProxyAPI/v6/internal/auth/gemini"
iflowauth "github.com/router-for-me/CLIProxyAPI/v6/internal/auth/iflow"
"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/kimi"
"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/qwen"
"github.com/router-for-me/CLIProxyAPI/v6/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v6/internal/misc"
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/internal/auth/antigravity"
"github.com/router-for-me/CLIProxyAPI/v7/internal/auth/claude"
"github.com/router-for-me/CLIProxyAPI/v7/internal/auth/codex"
geminiAuth "github.com/router-for-me/CLIProxyAPI/v7/internal/auth/gemini"
iflowauth "github.com/router-for-me/CLIProxyAPI/v7/internal/auth/iflow"
"github.com/router-for-me/CLIProxyAPI/v7/internal/auth/kimi"
"github.com/router-for-me/CLIProxyAPI/v7/internal/auth/qwen"
"github.com/router-for-me/CLIProxyAPI/v7/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v7/internal/misc"
"github.com/router-for-me/CLIProxyAPI/v7/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
log "github.com/sirupsen/logrus"
"github.com/tidwall/gjson"
"golang.org/x/oauth2"
@@ -390,6 +390,9 @@ func (h *Handler) buildAuthFileEntry(auth *coreauth.Auth) gin.H {
"source": "memory",
"size": int64(0),
}
entry["success"] = auth.Success
entry["failed"] = auth.Failed
entry["recent_requests"] = auth.RecentRequestsSnapshot(time.Now())
if email := authEmail(auth); email != "" {
entry["email"] = email
}
@@ -1055,6 +1058,17 @@ func (h *Handler) buildAuthFromFileData(path string, data []byte) (*coreauth.Aut
if hasLastRefresh {
auth.LastRefreshedAt = lastRefresh
}
if h != nil && h.authManager != nil {
if existing, ok := h.authManager.GetByID(authID); ok {
auth.CreatedAt = existing.CreatedAt
if !hasLastRefresh {
auth.LastRefreshedAt = existing.LastRefreshedAt
}
auth.NextRefreshAfter = existing.NextRefreshAfter
auth.Runtime = existing.Runtime
}
}
coreauth.ApplyCustomHeadersFromMetadata(auth)
return auth, hasLastRefresh, nil
}
@@ -1142,7 +1156,7 @@ func (h *Handler) PatchAuthFileStatus(c *gin.Context) {
c.JSON(http.StatusOK, gin.H{"status": "ok", "disabled": *req.Disabled})
}
// PatchAuthFileFields updates editable fields (prefix, proxy_url, priority, note) of an auth file.
// PatchAuthFileFields updates editable fields (prefix, proxy_url, headers, priority, note) of an auth file.
func (h *Handler) PatchAuthFileFields(c *gin.Context) {
if h.authManager == nil {
c.JSON(http.StatusServiceUnavailable, gin.H{"error": "core auth manager unavailable"})
@@ -1150,11 +1164,12 @@ func (h *Handler) PatchAuthFileFields(c *gin.Context) {
}
var req struct {
Name string `json:"name"`
Prefix *string `json:"prefix"`
ProxyURL *string `json:"proxy_url"`
Priority *int `json:"priority"`
Note *string `json:"note"`
Name string `json:"name"`
Prefix *string `json:"prefix"`
ProxyURL *string `json:"proxy_url"`
Headers map[string]string `json:"headers"`
Priority *int `json:"priority"`
Note *string `json:"note"`
}
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusBadRequest, gin.H{"error": "invalid request body"})
@@ -1190,39 +1205,117 @@ func (h *Handler) PatchAuthFileFields(c *gin.Context) {
changed := false
if req.Prefix != nil {
targetAuth.Prefix = *req.Prefix
prefix := strings.TrimSpace(*req.Prefix)
targetAuth.Prefix = prefix
if targetAuth.Metadata == nil {
targetAuth.Metadata = make(map[string]any)
}
if targetAuth.Attributes == nil {
targetAuth.Attributes = make(map[string]string)
}
if targetAuth.Prefix == "" {
if prefix == "" {
delete(targetAuth.Metadata, "prefix")
delete(targetAuth.Attributes, "prefix")
} else {
targetAuth.Metadata["prefix"] = targetAuth.Prefix
targetAuth.Attributes["prefix"] = targetAuth.Prefix
targetAuth.Metadata["prefix"] = prefix
targetAuth.Attributes["prefix"] = prefix
}
changed = true
}
if req.ProxyURL != nil {
targetAuth.ProxyURL = *req.ProxyURL
proxyURL := strings.TrimSpace(*req.ProxyURL)
targetAuth.ProxyURL = proxyURL
if targetAuth.Metadata == nil {
targetAuth.Metadata = make(map[string]any)
}
if targetAuth.Attributes == nil {
targetAuth.Attributes = make(map[string]string)
}
if targetAuth.ProxyURL == "" {
if proxyURL == "" {
delete(targetAuth.Metadata, "proxy_url")
delete(targetAuth.Attributes, "proxy_url")
} else {
targetAuth.Metadata["proxy_url"] = targetAuth.ProxyURL
targetAuth.Attributes["proxy_url"] = targetAuth.ProxyURL
targetAuth.Metadata["proxy_url"] = proxyURL
targetAuth.Attributes["proxy_url"] = proxyURL
}
changed = true
}
if len(req.Headers) > 0 {
existingHeaders := coreauth.ExtractCustomHeadersFromMetadata(targetAuth.Metadata)
nextHeaders := make(map[string]string, len(existingHeaders))
for k, v := range existingHeaders {
nextHeaders[k] = v
}
headerChanged := false
for key, value := range req.Headers {
name := strings.TrimSpace(key)
if name == "" {
continue
}
val := strings.TrimSpace(value)
attrKey := "header:" + name
if val == "" {
if _, ok := nextHeaders[name]; ok {
delete(nextHeaders, name)
headerChanged = true
}
if targetAuth.Attributes != nil {
if _, ok := targetAuth.Attributes[attrKey]; ok {
headerChanged = true
}
}
continue
}
if prev, ok := nextHeaders[name]; !ok || prev != val {
headerChanged = true
}
nextHeaders[name] = val
if targetAuth.Attributes != nil {
if prev, ok := targetAuth.Attributes[attrKey]; !ok || prev != val {
headerChanged = true
}
} else {
headerChanged = true
}
}
if headerChanged {
if targetAuth.Metadata == nil {
targetAuth.Metadata = make(map[string]any)
}
if targetAuth.Attributes == nil {
targetAuth.Attributes = make(map[string]string)
}
for key, value := range req.Headers {
name := strings.TrimSpace(key)
if name == "" {
continue
}
val := strings.TrimSpace(value)
attrKey := "header:" + name
if val == "" {
delete(nextHeaders, name)
delete(targetAuth.Attributes, attrKey)
continue
}
nextHeaders[name] = val
targetAuth.Attributes[attrKey] = val
}
if len(nextHeaders) == 0 {
delete(targetAuth.Metadata, "headers")
} else {
metaHeaders := make(map[string]any, len(nextHeaders))
for k, v := range nextHeaders {
metaHeaders[k] = v
}
targetAuth.Metadata["headers"] = metaHeaders
}
changed = true
}
}
if req.Priority != nil || req.Note != nil {
if targetAuth.Metadata == nil {
targetAuth.Metadata = make(map[string]any)
@@ -2092,10 +2185,8 @@ func (h *Handler) RequestQwenToken(c *gin.Context) {
fmt.Println("Initializing Qwen authentication...")
state := fmt.Sprintf("gem-%d", time.Now().UnixNano())
// Initialize Qwen auth service
qwenAuth := qwen.NewQwenAuth(h.cfg)
// Generate authorization URL
deviceFlow, err := qwenAuth.InitiateDeviceFlow(ctx)
if err != nil {
log.Errorf("Failed to generate authorization URL: %v", err)
@@ -2115,9 +2206,7 @@ func (h *Handler) RequestQwenToken(c *gin.Context) {
return
}
// Create token storage
tokenStorage := qwenAuth.CreateTokenStorage(tokenData)
tokenStorage.Email = fmt.Sprintf("%d", time.Now().UnixMilli())
record := &coreauth.Auth{
ID: fmt.Sprintf("qwen-%s.json", tokenStorage.Email),
@@ -2644,23 +2733,10 @@ func performGeminiCLISetup(ctx context.Context, httpClient *http.Client, storage
finalProjectID := projectID
if responseProjectID != "" {
if explicitProject && !strings.EqualFold(responseProjectID, projectID) {
// Check if this is a free user (gen-lang-client projects or free/legacy tier)
isFreeUser := strings.HasPrefix(projectID, "gen-lang-client-") ||
strings.EqualFold(tierID, "FREE") ||
strings.EqualFold(tierID, "LEGACY")
if isFreeUser {
// For free users, use backend project ID for preview model access
log.Infof("Gemini onboarding: frontend project %s maps to backend project %s", projectID, responseProjectID)
log.Infof("Using backend project ID: %s (recommended for preview model access)", responseProjectID)
finalProjectID = responseProjectID
} else {
// Pro users: keep requested project ID (original behavior)
log.Warnf("Gemini onboarding returned project %s instead of requested %s; keeping requested project ID.", responseProjectID, projectID)
}
} else {
finalProjectID = responseProjectID
log.Infof("Gemini onboarding: requested project %s maps to backend project %s", projectID, responseProjectID)
log.Infof("Using backend project ID: %s", responseProjectID)
}
finalProjectID = responseProjectID
}
storage.ProjectID = strings.TrimSpace(finalProjectID)
@@ -12,8 +12,8 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
func TestUploadAuthFile_BatchMultipart(t *testing.T) {
@@ -11,8 +11,8 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
func TestDeleteAuthFile_UsesAuthPathFromManager(t *testing.T) {
@@ -9,7 +9,7 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
func TestDownloadAuthFile_ReturnsFile(t *testing.T) {
@@ -11,7 +11,7 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
func TestDownloadAuthFile_PreventsWindowsSlashTraversal(t *testing.T) {
@@ -0,0 +1,164 @@
package management
import (
"context"
"encoding/json"
"net/http"
"net/http/httptest"
"strings"
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
func TestPatchAuthFileFields_MergeHeadersAndDeleteEmptyValues(t *testing.T) {
t.Setenv("MANAGEMENT_PASSWORD", "")
gin.SetMode(gin.TestMode)
store := &memoryAuthStore{}
manager := coreauth.NewManager(store, nil, nil)
record := &coreauth.Auth{
ID: "test.json",
FileName: "test.json",
Provider: "claude",
Attributes: map[string]string{
"path": "/tmp/test.json",
"header:X-Old": "old",
"header:X-Remove": "gone",
},
Metadata: map[string]any{
"type": "claude",
"headers": map[string]any{
"X-Old": "old",
"X-Remove": "gone",
},
},
}
if _, errRegister := manager.Register(context.Background(), record); errRegister != nil {
t.Fatalf("failed to register auth record: %v", errRegister)
}
h := NewHandlerWithoutConfigFilePath(&config.Config{AuthDir: t.TempDir()}, manager)
body := `{"name":"test.json","prefix":"p1","proxy_url":"http://proxy.local","headers":{"X-Old":"new","X-New":"v","X-Remove":" ","X-Nope":""}}`
rec := httptest.NewRecorder()
ctx, _ := gin.CreateTestContext(rec)
req := httptest.NewRequest(http.MethodPatch, "/v0/management/auth-files/fields", strings.NewReader(body))
req.Header.Set("Content-Type", "application/json")
ctx.Request = req
h.PatchAuthFileFields(ctx)
if rec.Code != http.StatusOK {
t.Fatalf("expected status %d, got %d with body %s", http.StatusOK, rec.Code, rec.Body.String())
}
updated, ok := manager.GetByID("test.json")
if !ok || updated == nil {
t.Fatalf("expected auth record to exist after patch")
}
if updated.Prefix != "p1" {
t.Fatalf("prefix = %q, want %q", updated.Prefix, "p1")
}
if updated.ProxyURL != "http://proxy.local" {
t.Fatalf("proxy_url = %q, want %q", updated.ProxyURL, "http://proxy.local")
}
if updated.Metadata == nil {
t.Fatalf("expected metadata to be non-nil")
}
if got, _ := updated.Metadata["prefix"].(string); got != "p1" {
t.Fatalf("metadata.prefix = %q, want %q", got, "p1")
}
if got, _ := updated.Metadata["proxy_url"].(string); got != "http://proxy.local" {
t.Fatalf("metadata.proxy_url = %q, want %q", got, "http://proxy.local")
}
headersMeta, ok := updated.Metadata["headers"].(map[string]any)
if !ok {
raw, _ := json.Marshal(updated.Metadata["headers"])
t.Fatalf("metadata.headers = %T (%s), want map[string]any", updated.Metadata["headers"], string(raw))
}
if got := headersMeta["X-Old"]; got != "new" {
t.Fatalf("metadata.headers.X-Old = %#v, want %q", got, "new")
}
if got := headersMeta["X-New"]; got != "v" {
t.Fatalf("metadata.headers.X-New = %#v, want %q", got, "v")
}
if _, ok := headersMeta["X-Remove"]; ok {
t.Fatalf("expected metadata.headers.X-Remove to be deleted")
}
if _, ok := headersMeta["X-Nope"]; ok {
t.Fatalf("expected metadata.headers.X-Nope to be absent")
}
if got := updated.Attributes["header:X-Old"]; got != "new" {
t.Fatalf("attrs header:X-Old = %q, want %q", got, "new")
}
if got := updated.Attributes["header:X-New"]; got != "v" {
t.Fatalf("attrs header:X-New = %q, want %q", got, "v")
}
if _, ok := updated.Attributes["header:X-Remove"]; ok {
t.Fatalf("expected attrs header:X-Remove to be deleted")
}
if _, ok := updated.Attributes["header:X-Nope"]; ok {
t.Fatalf("expected attrs header:X-Nope to be absent")
}
}
func TestPatchAuthFileFields_HeadersEmptyMapIsNoop(t *testing.T) {
t.Setenv("MANAGEMENT_PASSWORD", "")
gin.SetMode(gin.TestMode)
store := &memoryAuthStore{}
manager := coreauth.NewManager(store, nil, nil)
record := &coreauth.Auth{
ID: "noop.json",
FileName: "noop.json",
Provider: "claude",
Attributes: map[string]string{
"path": "/tmp/noop.json",
"header:X-Kee": "1",
},
Metadata: map[string]any{
"type": "claude",
"headers": map[string]any{
"X-Kee": "1",
},
},
}
if _, errRegister := manager.Register(context.Background(), record); errRegister != nil {
t.Fatalf("failed to register auth record: %v", errRegister)
}
h := NewHandlerWithoutConfigFilePath(&config.Config{AuthDir: t.TempDir()}, manager)
body := `{"name":"noop.json","note":"hello","headers":{}}`
rec := httptest.NewRecorder()
ctx, _ := gin.CreateTestContext(rec)
req := httptest.NewRequest(http.MethodPatch, "/v0/management/auth-files/fields", strings.NewReader(body))
req.Header.Set("Content-Type", "application/json")
ctx.Request = req
h.PatchAuthFileFields(ctx)
if rec.Code != http.StatusOK {
t.Fatalf("expected status %d, got %d with body %s", http.StatusOK, rec.Code, rec.Body.String())
}
updated, ok := manager.GetByID("noop.json")
if !ok || updated == nil {
t.Fatalf("expected auth record to exist after patch")
}
if got := updated.Attributes["header:X-Kee"]; got != "1" {
t.Fatalf("attrs header:X-Kee = %q, want %q", got, "1")
}
headersMeta, ok := updated.Metadata["headers"].(map[string]any)
if !ok {
t.Fatalf("expected metadata.headers to remain a map, got %T", updated.Metadata["headers"])
}
if got := headersMeta["X-Kee"]; got != "1" {
t.Fatalf("metadata.headers.X-Kee = %#v, want %q", got, "1")
}
}
@@ -0,0 +1,94 @@
package management
import (
"context"
"encoding/json"
"net/http"
"net/http/httptest"
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
func TestListAuthFiles_IncludesRecentRequestsBuckets(t *testing.T) {
t.Setenv("MANAGEMENT_PASSWORD", "")
gin.SetMode(gin.TestMode)
manager := coreauth.NewManager(nil, nil, nil)
record := &coreauth.Auth{
ID: "runtime-only-auth-1",
Provider: "codex",
Attributes: map[string]string{
"runtime_only": "true",
},
Metadata: map[string]any{
"type": "codex",
},
}
if _, errRegister := manager.Register(context.Background(), record); errRegister != nil {
t.Fatalf("failed to register auth record: %v", errRegister)
}
h := NewHandlerWithoutConfigFilePath(&config.Config{AuthDir: t.TempDir()}, manager)
h.tokenStore = &memoryAuthStore{}
rec := httptest.NewRecorder()
ginCtx, _ := gin.CreateTestContext(rec)
req := httptest.NewRequest(http.MethodGet, "/v0/management/auth-files", nil)
ginCtx.Request = req
h.ListAuthFiles(ginCtx)
if rec.Code != http.StatusOK {
t.Fatalf("expected list status %d, got %d with body %s", http.StatusOK, rec.Code, rec.Body.String())
}
var payload map[string]any
if errUnmarshal := json.Unmarshal(rec.Body.Bytes(), &payload); errUnmarshal != nil {
t.Fatalf("failed to decode list payload: %v", errUnmarshal)
}
filesRaw, ok := payload["files"].([]any)
if !ok {
t.Fatalf("expected files array, payload: %#v", payload)
}
if len(filesRaw) != 1 {
t.Fatalf("expected 1 auth entry, got %d", len(filesRaw))
}
fileEntry, ok := filesRaw[0].(map[string]any)
if !ok {
t.Fatalf("expected file entry object, got %#v", filesRaw[0])
}
if _, ok := fileEntry["success"].(float64); !ok {
t.Fatalf("expected success number, got %#v", fileEntry["success"])
}
if _, ok := fileEntry["failed"].(float64); !ok {
t.Fatalf("expected failed number, got %#v", fileEntry["failed"])
}
recentRaw, ok := fileEntry["recent_requests"].([]any)
if !ok {
t.Fatalf("expected recent_requests array, got %#v", fileEntry["recent_requests"])
}
if len(recentRaw) != 20 {
t.Fatalf("expected 20 recent_requests buckets, got %d", len(recentRaw))
}
for idx, item := range recentRaw {
bucket, ok := item.(map[string]any)
if !ok {
t.Fatalf("expected bucket object at %d, got %#v", idx, item)
}
if _, ok := bucket["time"].(string); !ok {
t.Fatalf("expected bucket time string at %d, got %#v", idx, bucket["time"])
}
if _, ok := bucket["success"].(float64); !ok {
t.Fatalf("expected bucket success number at %d, got %#v", idx, bucket["success"])
}
if _, ok := bucket["failed"].(float64); !ok {
t.Fatalf("expected bucket failed number at %d, got %#v", idx, bucket["failed"])
}
}
}
@@ -6,8 +6,8 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/internal/registry"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
type authResetCooldownRequest struct {
@@ -4,7 +4,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/cache"
"github.com/router-for-me/CLIProxyAPI/v7/internal/cache"
)
// CacheStatsResponse represents the response for cache statistics.
@@ -0,0 +1,243 @@
package management
import (
"fmt"
"strings"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/watcher/synthesizer"
)
type geminiKeyWithAuthIndex struct {
config.GeminiKey
AuthIndex string `json:"auth-index,omitempty"`
}
type claudeKeyWithAuthIndex struct {
config.ClaudeKey
AuthIndex string `json:"auth-index,omitempty"`
}
type codexKeyWithAuthIndex struct {
config.CodexKey
AuthIndex string `json:"auth-index,omitempty"`
}
type vertexCompatKeyWithAuthIndex struct {
config.VertexCompatKey
AuthIndex string `json:"auth-index,omitempty"`
}
type openAICompatibilityAPIKeyWithAuthIndex struct {
config.OpenAICompatibilityAPIKey
AuthIndex string `json:"auth-index,omitempty"`
}
type openAICompatibilityWithAuthIndex struct {
Name string `json:"name"`
Priority int `json:"priority,omitempty"`
Disabled bool `json:"disabled"`
Prefix string `json:"prefix,omitempty"`
BaseURL string `json:"base-url"`
APIKeyEntries []openAICompatibilityAPIKeyWithAuthIndex `json:"api-key-entries,omitempty"`
Models []config.OpenAICompatibilityModel `json:"models,omitempty"`
Headers map[string]string `json:"headers,omitempty"`
AuthIndex string `json:"auth-index,omitempty"`
}
func (h *Handler) liveAuthIndexByID() map[string]string {
out := map[string]string{}
if h == nil {
return out
}
h.mu.Lock()
manager := h.authManager
h.mu.Unlock()
if manager == nil {
return out
}
// authManager.List() returns clones, so EnsureIndex only affects these copies.
for _, auth := range manager.List() {
if auth == nil {
continue
}
id := strings.TrimSpace(auth.ID)
if id == "" {
continue
}
idx := strings.TrimSpace(auth.Index)
if idx == "" {
idx = auth.EnsureIndex()
}
if idx == "" {
continue
}
out[id] = idx
}
return out
}
func (h *Handler) geminiKeysWithAuthIndex() []geminiKeyWithAuthIndex {
if h == nil {
return nil
}
liveIndexByID := h.liveAuthIndexByID()
h.mu.Lock()
defer h.mu.Unlock()
if h.cfg == nil {
return nil
}
idGen := synthesizer.NewStableIDGenerator()
out := make([]geminiKeyWithAuthIndex, len(h.cfg.GeminiKey))
for i := range h.cfg.GeminiKey {
entry := h.cfg.GeminiKey[i]
authIndex := ""
if key := strings.TrimSpace(entry.APIKey); key != "" {
id, _ := idGen.Next("gemini:apikey", key, entry.BaseURL)
authIndex = liveIndexByID[id]
}
out[i] = geminiKeyWithAuthIndex{
GeminiKey: entry,
AuthIndex: authIndex,
}
}
return out
}
func (h *Handler) claudeKeysWithAuthIndex() []claudeKeyWithAuthIndex {
if h == nil {
return nil
}
liveIndexByID := h.liveAuthIndexByID()
h.mu.Lock()
defer h.mu.Unlock()
if h.cfg == nil {
return nil
}
idGen := synthesizer.NewStableIDGenerator()
out := make([]claudeKeyWithAuthIndex, len(h.cfg.ClaudeKey))
for i := range h.cfg.ClaudeKey {
entry := h.cfg.ClaudeKey[i]
authIndex := ""
if key := strings.TrimSpace(entry.APIKey); key != "" {
id, _ := idGen.Next("claude:apikey", key, entry.BaseURL)
authIndex = liveIndexByID[id]
}
out[i] = claudeKeyWithAuthIndex{
ClaudeKey: entry,
AuthIndex: authIndex,
}
}
return out
}
func (h *Handler) codexKeysWithAuthIndex() []codexKeyWithAuthIndex {
if h == nil {
return nil
}
liveIndexByID := h.liveAuthIndexByID()
h.mu.Lock()
defer h.mu.Unlock()
if h.cfg == nil {
return nil
}
idGen := synthesizer.NewStableIDGenerator()
out := make([]codexKeyWithAuthIndex, len(h.cfg.CodexKey))
for i := range h.cfg.CodexKey {
entry := h.cfg.CodexKey[i]
authIndex := ""
if key := strings.TrimSpace(entry.APIKey); key != "" {
id, _ := idGen.Next("codex:apikey", key, entry.BaseURL)
authIndex = liveIndexByID[id]
}
out[i] = codexKeyWithAuthIndex{
CodexKey: entry,
AuthIndex: authIndex,
}
}
return out
}
func (h *Handler) vertexCompatKeysWithAuthIndex() []vertexCompatKeyWithAuthIndex {
if h == nil {
return nil
}
liveIndexByID := h.liveAuthIndexByID()
h.mu.Lock()
defer h.mu.Unlock()
if h.cfg == nil {
return nil
}
idGen := synthesizer.NewStableIDGenerator()
out := make([]vertexCompatKeyWithAuthIndex, len(h.cfg.VertexCompatAPIKey))
for i := range h.cfg.VertexCompatAPIKey {
entry := h.cfg.VertexCompatAPIKey[i]
id, _ := idGen.Next("vertex:apikey", entry.APIKey, entry.BaseURL, entry.ProxyURL)
authIndex := liveIndexByID[id]
out[i] = vertexCompatKeyWithAuthIndex{
VertexCompatKey: entry,
AuthIndex: authIndex,
}
}
return out
}
func (h *Handler) openAICompatibilityWithAuthIndex() []openAICompatibilityWithAuthIndex {
if h == nil {
return nil
}
liveIndexByID := h.liveAuthIndexByID()
h.mu.Lock()
defer h.mu.Unlock()
if h.cfg == nil {
return nil
}
normalized := normalizedOpenAICompatibilityEntries(h.cfg.OpenAICompatibility)
out := make([]openAICompatibilityWithAuthIndex, len(normalized))
idGen := synthesizer.NewStableIDGenerator()
for i := range normalized {
entry := normalized[i]
providerName := strings.ToLower(strings.TrimSpace(entry.Name))
if providerName == "" {
providerName = "openai-compatibility"
}
idKind := fmt.Sprintf("openai-compatibility:%s", providerName)
response := openAICompatibilityWithAuthIndex{
Name: entry.Name,
Priority: entry.Priority,
Disabled: entry.Disabled,
Prefix: entry.Prefix,
BaseURL: entry.BaseURL,
Models: entry.Models,
Headers: entry.Headers,
AuthIndex: "",
}
if len(entry.APIKeyEntries) == 0 {
id, _ := idGen.Next(idKind, entry.BaseURL)
response.AuthIndex = liveIndexByID[id]
} else {
response.APIKeyEntries = make([]openAICompatibilityAPIKeyWithAuthIndex, len(entry.APIKeyEntries))
for j := range entry.APIKeyEntries {
apiKeyEntry := entry.APIKeyEntries[j]
id, _ := idGen.Next(idKind, apiKeyEntry.APIKey, entry.BaseURL, apiKeyEntry.ProxyURL)
response.APIKeyEntries[j] = openAICompatibilityAPIKeyWithAuthIndex{
OpenAICompatibilityAPIKey: apiKeyEntry,
AuthIndex: liveIndexByID[id],
}
}
}
out[i] = response
}
return out
}
@@ -11,9 +11,9 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
sdkconfig "github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
sdkconfig "github.com/router-for-me/CLIProxyAPI/v7/sdk/config"
log "github.com/sirupsen/logrus"
"gopkg.in/yaml.v3"
)
+193 -52
View File
@@ -6,7 +6,7 @@ import (
"strings"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
// Generic helpers for list[string]
@@ -120,7 +120,7 @@ func (h *Handler) DeleteAPIKeys(c *gin.Context) {
// gemini-api-key: []GeminiKey
func (h *Handler) GetGeminiKeys(c *gin.Context) {
c.JSON(200, gin.H{"gemini-api-key": h.cfg.GeminiKey})
c.JSON(200, gin.H{"gemini-api-key": h.geminiKeysWithAuthIndex()})
}
func (h *Handler) PutGeminiKeys(c *gin.Context) {
data, err := c.GetRawData()
@@ -139,9 +139,11 @@ func (h *Handler) PutGeminiKeys(c *gin.Context) {
}
arr = obj.Items
}
h.mu.Lock()
defer h.mu.Unlock()
h.cfg.GeminiKey = append([]config.GeminiKey(nil), arr...)
h.cfg.SanitizeGeminiKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) PatchGeminiKey(c *gin.Context) {
type geminiKeyPatch struct {
@@ -161,6 +163,9 @@ func (h *Handler) PatchGeminiKey(c *gin.Context) {
c.JSON(400, gin.H{"error": "invalid body"})
return
}
h.mu.Lock()
defer h.mu.Unlock()
targetIndex := -1
if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.GeminiKey) {
targetIndex = *body.Index
@@ -187,7 +192,7 @@ func (h *Handler) PatchGeminiKey(c *gin.Context) {
if trimmed == "" {
h.cfg.GeminiKey = append(h.cfg.GeminiKey[:targetIndex], h.cfg.GeminiKey[targetIndex+1:]...)
h.cfg.SanitizeGeminiKeys()
h.persist(c)
h.persistLocked(c)
return
}
entry.APIKey = trimmed
@@ -209,24 +214,53 @@ func (h *Handler) PatchGeminiKey(c *gin.Context) {
}
h.cfg.GeminiKey[targetIndex] = entry
h.cfg.SanitizeGeminiKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) DeleteGeminiKey(c *gin.Context) {
h.mu.Lock()
defer h.mu.Unlock()
if val := strings.TrimSpace(c.Query("api-key")); val != "" {
out := make([]config.GeminiKey, 0, len(h.cfg.GeminiKey))
for _, v := range h.cfg.GeminiKey {
if v.APIKey != val {
if baseRaw, okBase := c.GetQuery("base-url"); okBase {
base := strings.TrimSpace(baseRaw)
out := make([]config.GeminiKey, 0, len(h.cfg.GeminiKey))
for _, v := range h.cfg.GeminiKey {
if strings.TrimSpace(v.APIKey) == val && strings.TrimSpace(v.BaseURL) == base {
continue
}
out = append(out, v)
}
if len(out) != len(h.cfg.GeminiKey) {
h.cfg.GeminiKey = out
h.cfg.SanitizeGeminiKeys()
h.persistLocked(c)
} else {
c.JSON(404, gin.H{"error": "item not found"})
}
return
}
if len(out) != len(h.cfg.GeminiKey) {
h.cfg.GeminiKey = out
h.cfg.SanitizeGeminiKeys()
h.persist(c)
} else {
matchIndex := -1
matchCount := 0
for i := range h.cfg.GeminiKey {
if strings.TrimSpace(h.cfg.GeminiKey[i].APIKey) == val {
matchCount++
if matchIndex == -1 {
matchIndex = i
}
}
}
if matchCount == 0 {
c.JSON(404, gin.H{"error": "item not found"})
return
}
if matchCount > 1 {
c.JSON(400, gin.H{"error": "multiple items match api-key; base-url is required"})
return
}
h.cfg.GeminiKey = append(h.cfg.GeminiKey[:matchIndex], h.cfg.GeminiKey[matchIndex+1:]...)
h.cfg.SanitizeGeminiKeys()
h.persistLocked(c)
return
}
if idxStr := c.Query("index"); idxStr != "" {
@@ -234,7 +268,7 @@ func (h *Handler) DeleteGeminiKey(c *gin.Context) {
if _, err := fmt.Sscanf(idxStr, "%d", &idx); err == nil && idx >= 0 && idx < len(h.cfg.GeminiKey) {
h.cfg.GeminiKey = append(h.cfg.GeminiKey[:idx], h.cfg.GeminiKey[idx+1:]...)
h.cfg.SanitizeGeminiKeys()
h.persist(c)
h.persistLocked(c)
return
}
}
@@ -243,7 +277,7 @@ func (h *Handler) DeleteGeminiKey(c *gin.Context) {
// claude-api-key: []ClaudeKey
func (h *Handler) GetClaudeKeys(c *gin.Context) {
c.JSON(200, gin.H{"claude-api-key": h.cfg.ClaudeKey})
c.JSON(200, gin.H{"claude-api-key": h.claudeKeysWithAuthIndex()})
}
func (h *Handler) PutClaudeKeys(c *gin.Context) {
data, err := c.GetRawData()
@@ -265,9 +299,11 @@ func (h *Handler) PutClaudeKeys(c *gin.Context) {
for i := range arr {
normalizeClaudeKey(&arr[i])
}
h.mu.Lock()
defer h.mu.Unlock()
h.cfg.ClaudeKey = arr
h.cfg.SanitizeClaudeKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) PatchClaudeKey(c *gin.Context) {
type claudeKeyPatch struct {
@@ -288,6 +324,9 @@ func (h *Handler) PatchClaudeKey(c *gin.Context) {
c.JSON(400, gin.H{"error": "invalid body"})
return
}
h.mu.Lock()
defer h.mu.Unlock()
targetIndex := -1
if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.ClaudeKey) {
targetIndex = *body.Index
@@ -331,20 +370,47 @@ func (h *Handler) PatchClaudeKey(c *gin.Context) {
normalizeClaudeKey(&entry)
h.cfg.ClaudeKey[targetIndex] = entry
h.cfg.SanitizeClaudeKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) DeleteClaudeKey(c *gin.Context) {
if val := c.Query("api-key"); val != "" {
out := make([]config.ClaudeKey, 0, len(h.cfg.ClaudeKey))
for _, v := range h.cfg.ClaudeKey {
if v.APIKey != val {
h.mu.Lock()
defer h.mu.Unlock()
if val := strings.TrimSpace(c.Query("api-key")); val != "" {
if baseRaw, okBase := c.GetQuery("base-url"); okBase {
base := strings.TrimSpace(baseRaw)
out := make([]config.ClaudeKey, 0, len(h.cfg.ClaudeKey))
for _, v := range h.cfg.ClaudeKey {
if strings.TrimSpace(v.APIKey) == val && strings.TrimSpace(v.BaseURL) == base {
continue
}
out = append(out, v)
}
h.cfg.ClaudeKey = out
h.cfg.SanitizeClaudeKeys()
h.persistLocked(c)
return
}
matchIndex := -1
matchCount := 0
for i := range h.cfg.ClaudeKey {
if strings.TrimSpace(h.cfg.ClaudeKey[i].APIKey) == val {
matchCount++
if matchIndex == -1 {
matchIndex = i
}
}
}
if matchCount > 1 {
c.JSON(400, gin.H{"error": "multiple items match api-key; base-url is required"})
return
}
if matchIndex != -1 {
h.cfg.ClaudeKey = append(h.cfg.ClaudeKey[:matchIndex], h.cfg.ClaudeKey[matchIndex+1:]...)
}
h.cfg.ClaudeKey = out
h.cfg.SanitizeClaudeKeys()
h.persist(c)
h.persistLocked(c)
return
}
if idxStr := c.Query("index"); idxStr != "" {
@@ -353,7 +419,7 @@ func (h *Handler) DeleteClaudeKey(c *gin.Context) {
if err == nil && idx >= 0 && idx < len(h.cfg.ClaudeKey) {
h.cfg.ClaudeKey = append(h.cfg.ClaudeKey[:idx], h.cfg.ClaudeKey[idx+1:]...)
h.cfg.SanitizeClaudeKeys()
h.persist(c)
h.persistLocked(c)
return
}
}
@@ -362,7 +428,7 @@ func (h *Handler) DeleteClaudeKey(c *gin.Context) {
// openai-compatibility: []OpenAICompatibility
func (h *Handler) GetOpenAICompat(c *gin.Context) {
c.JSON(200, gin.H{"openai-compatibility": normalizedOpenAICompatibilityEntries(h.cfg.OpenAICompatibility)})
c.JSON(200, gin.H{"openai-compatibility": h.openAICompatibilityWithAuthIndex()})
}
func (h *Handler) PutOpenAICompat(c *gin.Context) {
data, err := c.GetRawData()
@@ -388,14 +454,17 @@ func (h *Handler) PutOpenAICompat(c *gin.Context) {
filtered = append(filtered, arr[i])
}
}
h.mu.Lock()
defer h.mu.Unlock()
h.cfg.OpenAICompatibility = filtered
h.cfg.SanitizeOpenAICompatibility()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) PatchOpenAICompat(c *gin.Context) {
type openAICompatPatch struct {
Name *string `json:"name"`
Prefix *string `json:"prefix"`
Disabled *bool `json:"disabled"`
BaseURL *string `json:"base-url"`
APIKeyEntries *[]config.OpenAICompatibilityAPIKey `json:"api-key-entries"`
Models *[]config.OpenAICompatibilityModel `json:"models"`
@@ -410,6 +479,9 @@ func (h *Handler) PatchOpenAICompat(c *gin.Context) {
c.JSON(400, gin.H{"error": "invalid body"})
return
}
h.mu.Lock()
defer h.mu.Unlock()
targetIndex := -1
if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.OpenAICompatibility) {
targetIndex = *body.Index
@@ -435,12 +507,15 @@ func (h *Handler) PatchOpenAICompat(c *gin.Context) {
if body.Value.Prefix != nil {
entry.Prefix = strings.TrimSpace(*body.Value.Prefix)
}
if body.Value.Disabled != nil {
entry.Disabled = *body.Value.Disabled
}
if body.Value.BaseURL != nil {
trimmed := strings.TrimSpace(*body.Value.BaseURL)
if trimmed == "" {
h.cfg.OpenAICompatibility = append(h.cfg.OpenAICompatibility[:targetIndex], h.cfg.OpenAICompatibility[targetIndex+1:]...)
h.cfg.SanitizeOpenAICompatibility()
h.persist(c)
h.persistLocked(c)
return
}
entry.BaseURL = trimmed
@@ -457,10 +532,12 @@ func (h *Handler) PatchOpenAICompat(c *gin.Context) {
normalizeOpenAICompatibilityEntry(&entry)
h.cfg.OpenAICompatibility[targetIndex] = entry
h.cfg.SanitizeOpenAICompatibility()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) DeleteOpenAICompat(c *gin.Context) {
h.mu.Lock()
defer h.mu.Unlock()
if name := c.Query("name"); name != "" {
out := make([]config.OpenAICompatibility, 0, len(h.cfg.OpenAICompatibility))
for _, v := range h.cfg.OpenAICompatibility {
@@ -470,7 +547,7 @@ func (h *Handler) DeleteOpenAICompat(c *gin.Context) {
}
h.cfg.OpenAICompatibility = out
h.cfg.SanitizeOpenAICompatibility()
h.persist(c)
h.persistLocked(c)
return
}
if idxStr := c.Query("index"); idxStr != "" {
@@ -479,7 +556,7 @@ func (h *Handler) DeleteOpenAICompat(c *gin.Context) {
if err == nil && idx >= 0 && idx < len(h.cfg.OpenAICompatibility) {
h.cfg.OpenAICompatibility = append(h.cfg.OpenAICompatibility[:idx], h.cfg.OpenAICompatibility[idx+1:]...)
h.cfg.SanitizeOpenAICompatibility()
h.persist(c)
h.persistLocked(c)
return
}
}
@@ -488,7 +565,7 @@ func (h *Handler) DeleteOpenAICompat(c *gin.Context) {
// vertex-api-key: []VertexCompatKey
func (h *Handler) GetVertexCompatKeys(c *gin.Context) {
c.JSON(200, gin.H{"vertex-api-key": h.cfg.VertexCompatAPIKey})
c.JSON(200, gin.H{"vertex-api-key": h.vertexCompatKeysWithAuthIndex()})
}
func (h *Handler) PutVertexCompatKeys(c *gin.Context) {
data, err := c.GetRawData()
@@ -514,9 +591,11 @@ func (h *Handler) PutVertexCompatKeys(c *gin.Context) {
return
}
}
h.mu.Lock()
defer h.mu.Unlock()
h.cfg.VertexCompatAPIKey = append([]config.VertexCompatKey(nil), arr...)
h.cfg.SanitizeVertexCompatKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) PatchVertexCompatKey(c *gin.Context) {
type vertexCompatPatch struct {
@@ -537,6 +616,9 @@ func (h *Handler) PatchVertexCompatKey(c *gin.Context) {
c.JSON(400, gin.H{"error": "invalid body"})
return
}
h.mu.Lock()
defer h.mu.Unlock()
targetIndex := -1
if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.VertexCompatAPIKey) {
targetIndex = *body.Index
@@ -563,7 +645,7 @@ func (h *Handler) PatchVertexCompatKey(c *gin.Context) {
if trimmed == "" {
h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:targetIndex], h.cfg.VertexCompatAPIKey[targetIndex+1:]...)
h.cfg.SanitizeVertexCompatKeys()
h.persist(c)
h.persistLocked(c)
return
}
entry.APIKey = trimmed
@@ -576,7 +658,7 @@ func (h *Handler) PatchVertexCompatKey(c *gin.Context) {
if trimmed == "" {
h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:targetIndex], h.cfg.VertexCompatAPIKey[targetIndex+1:]...)
h.cfg.SanitizeVertexCompatKeys()
h.persist(c)
h.persistLocked(c)
return
}
entry.BaseURL = trimmed
@@ -596,20 +678,47 @@ func (h *Handler) PatchVertexCompatKey(c *gin.Context) {
normalizeVertexCompatKey(&entry)
h.cfg.VertexCompatAPIKey[targetIndex] = entry
h.cfg.SanitizeVertexCompatKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) DeleteVertexCompatKey(c *gin.Context) {
h.mu.Lock()
defer h.mu.Unlock()
if val := strings.TrimSpace(c.Query("api-key")); val != "" {
out := make([]config.VertexCompatKey, 0, len(h.cfg.VertexCompatAPIKey))
for _, v := range h.cfg.VertexCompatAPIKey {
if v.APIKey != val {
if baseRaw, okBase := c.GetQuery("base-url"); okBase {
base := strings.TrimSpace(baseRaw)
out := make([]config.VertexCompatKey, 0, len(h.cfg.VertexCompatAPIKey))
for _, v := range h.cfg.VertexCompatAPIKey {
if strings.TrimSpace(v.APIKey) == val && strings.TrimSpace(v.BaseURL) == base {
continue
}
out = append(out, v)
}
h.cfg.VertexCompatAPIKey = out
h.cfg.SanitizeVertexCompatKeys()
h.persistLocked(c)
return
}
matchIndex := -1
matchCount := 0
for i := range h.cfg.VertexCompatAPIKey {
if strings.TrimSpace(h.cfg.VertexCompatAPIKey[i].APIKey) == val {
matchCount++
if matchIndex == -1 {
matchIndex = i
}
}
}
if matchCount > 1 {
c.JSON(400, gin.H{"error": "multiple items match api-key; base-url is required"})
return
}
if matchIndex != -1 {
h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:matchIndex], h.cfg.VertexCompatAPIKey[matchIndex+1:]...)
}
h.cfg.VertexCompatAPIKey = out
h.cfg.SanitizeVertexCompatKeys()
h.persist(c)
h.persistLocked(c)
return
}
if idxStr := c.Query("index"); idxStr != "" {
@@ -618,7 +727,7 @@ func (h *Handler) DeleteVertexCompatKey(c *gin.Context) {
if errScan == nil && idx >= 0 && idx < len(h.cfg.VertexCompatAPIKey) {
h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:idx], h.cfg.VertexCompatAPIKey[idx+1:]...)
h.cfg.SanitizeVertexCompatKeys()
h.persist(c)
h.persistLocked(c)
return
}
}
@@ -809,7 +918,7 @@ func (h *Handler) DeleteOAuthModelAlias(c *gin.Context) {
// codex-api-key: []CodexKey
func (h *Handler) GetCodexKeys(c *gin.Context) {
c.JSON(200, gin.H{"codex-api-key": h.cfg.CodexKey})
c.JSON(200, gin.H{"codex-api-key": h.codexKeysWithAuthIndex()})
}
func (h *Handler) PutCodexKeys(c *gin.Context) {
data, err := c.GetRawData()
@@ -838,9 +947,11 @@ func (h *Handler) PutCodexKeys(c *gin.Context) {
}
filtered = append(filtered, entry)
}
h.mu.Lock()
defer h.mu.Unlock()
h.cfg.CodexKey = filtered
h.cfg.SanitizeCodexKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) PatchCodexKey(c *gin.Context) {
type codexKeyPatch struct {
@@ -861,6 +972,9 @@ func (h *Handler) PatchCodexKey(c *gin.Context) {
c.JSON(400, gin.H{"error": "invalid body"})
return
}
h.mu.Lock()
defer h.mu.Unlock()
targetIndex := -1
if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.CodexKey) {
targetIndex = *body.Index
@@ -891,7 +1005,7 @@ func (h *Handler) PatchCodexKey(c *gin.Context) {
if trimmed == "" {
h.cfg.CodexKey = append(h.cfg.CodexKey[:targetIndex], h.cfg.CodexKey[targetIndex+1:]...)
h.cfg.SanitizeCodexKeys()
h.persist(c)
h.persistLocked(c)
return
}
entry.BaseURL = trimmed
@@ -911,20 +1025,47 @@ func (h *Handler) PatchCodexKey(c *gin.Context) {
normalizeCodexKey(&entry)
h.cfg.CodexKey[targetIndex] = entry
h.cfg.SanitizeCodexKeys()
h.persist(c)
h.persistLocked(c)
}
func (h *Handler) DeleteCodexKey(c *gin.Context) {
if val := c.Query("api-key"); val != "" {
out := make([]config.CodexKey, 0, len(h.cfg.CodexKey))
for _, v := range h.cfg.CodexKey {
if v.APIKey != val {
h.mu.Lock()
defer h.mu.Unlock()
if val := strings.TrimSpace(c.Query("api-key")); val != "" {
if baseRaw, okBase := c.GetQuery("base-url"); okBase {
base := strings.TrimSpace(baseRaw)
out := make([]config.CodexKey, 0, len(h.cfg.CodexKey))
for _, v := range h.cfg.CodexKey {
if strings.TrimSpace(v.APIKey) == val && strings.TrimSpace(v.BaseURL) == base {
continue
}
out = append(out, v)
}
h.cfg.CodexKey = out
h.cfg.SanitizeCodexKeys()
h.persistLocked(c)
return
}
matchIndex := -1
matchCount := 0
for i := range h.cfg.CodexKey {
if strings.TrimSpace(h.cfg.CodexKey[i].APIKey) == val {
matchCount++
if matchIndex == -1 {
matchIndex = i
}
}
}
if matchCount > 1 {
c.JSON(400, gin.H{"error": "multiple items match api-key; base-url is required"})
return
}
if matchIndex != -1 {
h.cfg.CodexKey = append(h.cfg.CodexKey[:matchIndex], h.cfg.CodexKey[matchIndex+1:]...)
}
h.cfg.CodexKey = out
h.cfg.SanitizeCodexKeys()
h.persist(c)
h.persistLocked(c)
return
}
if idxStr := c.Query("index"); idxStr != "" {
@@ -933,7 +1074,7 @@ func (h *Handler) DeleteCodexKey(c *gin.Context) {
if err == nil && idx >= 0 && idx < len(h.cfg.CodexKey) {
h.cfg.CodexKey = append(h.cfg.CodexKey[:idx], h.cfg.CodexKey[idx+1:]...)
h.cfg.SanitizeCodexKeys()
h.persist(c)
h.persistLocked(c)
return
}
}
@@ -0,0 +1,172 @@
package management
import (
"net/http"
"net/http/httptest"
"os"
"path/filepath"
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
func writeTestConfigFile(t *testing.T) string {
t.Helper()
dir := t.TempDir()
path := filepath.Join(dir, "config.yaml")
if errWrite := os.WriteFile(path, []byte("{}\n"), 0o600); errWrite != nil {
t.Fatalf("failed to write test config: %v", errWrite)
}
return path
}
func TestDeleteGeminiKey_RequiresBaseURLWhenAPIKeyDuplicated(t *testing.T) {
t.Parallel()
gin.SetMode(gin.TestMode)
h := &Handler{
cfg: &config.Config{
GeminiKey: []config.GeminiKey{
{APIKey: "shared-key", BaseURL: "https://a.example.com"},
{APIKey: "shared-key", BaseURL: "https://b.example.com"},
},
},
configFilePath: writeTestConfigFile(t),
}
rec := httptest.NewRecorder()
c, _ := gin.CreateTestContext(rec)
c.Request = httptest.NewRequest(http.MethodDelete, "/v0/management/gemini-api-key?api-key=shared-key", nil)
h.DeleteGeminiKey(c)
if rec.Code != http.StatusBadRequest {
t.Fatalf("status = %d, want %d; body=%s", rec.Code, http.StatusBadRequest, rec.Body.String())
}
if got := len(h.cfg.GeminiKey); got != 2 {
t.Fatalf("gemini keys len = %d, want 2", got)
}
}
func TestDeleteGeminiKey_DeletesOnlyMatchingBaseURL(t *testing.T) {
t.Parallel()
gin.SetMode(gin.TestMode)
h := &Handler{
cfg: &config.Config{
GeminiKey: []config.GeminiKey{
{APIKey: "shared-key", BaseURL: "https://a.example.com"},
{APIKey: "shared-key", BaseURL: "https://b.example.com"},
},
},
configFilePath: writeTestConfigFile(t),
}
rec := httptest.NewRecorder()
c, _ := gin.CreateTestContext(rec)
c.Request = httptest.NewRequest(http.MethodDelete, "/v0/management/gemini-api-key?api-key=shared-key&base-url=https://a.example.com", nil)
h.DeleteGeminiKey(c)
if rec.Code != http.StatusOK {
t.Fatalf("status = %d, want %d; body=%s", rec.Code, http.StatusOK, rec.Body.String())
}
if got := len(h.cfg.GeminiKey); got != 1 {
t.Fatalf("gemini keys len = %d, want 1", got)
}
if got := h.cfg.GeminiKey[0].BaseURL; got != "https://b.example.com" {
t.Fatalf("remaining base-url = %q, want %q", got, "https://b.example.com")
}
}
func TestDeleteClaudeKey_DeletesEmptyBaseURLWhenExplicitlyProvided(t *testing.T) {
t.Parallel()
gin.SetMode(gin.TestMode)
h := &Handler{
cfg: &config.Config{
ClaudeKey: []config.ClaudeKey{
{APIKey: "shared-key", BaseURL: ""},
{APIKey: "shared-key", BaseURL: "https://claude.example.com"},
},
},
configFilePath: writeTestConfigFile(t),
}
rec := httptest.NewRecorder()
c, _ := gin.CreateTestContext(rec)
c.Request = httptest.NewRequest(http.MethodDelete, "/v0/management/claude-api-key?api-key=shared-key&base-url=", nil)
h.DeleteClaudeKey(c)
if rec.Code != http.StatusOK {
t.Fatalf("status = %d, want %d; body=%s", rec.Code, http.StatusOK, rec.Body.String())
}
if got := len(h.cfg.ClaudeKey); got != 1 {
t.Fatalf("claude keys len = %d, want 1", got)
}
if got := h.cfg.ClaudeKey[0].BaseURL; got != "https://claude.example.com" {
t.Fatalf("remaining base-url = %q, want %q", got, "https://claude.example.com")
}
}
func TestDeleteVertexCompatKey_DeletesOnlyMatchingBaseURL(t *testing.T) {
t.Parallel()
gin.SetMode(gin.TestMode)
h := &Handler{
cfg: &config.Config{
VertexCompatAPIKey: []config.VertexCompatKey{
{APIKey: "shared-key", BaseURL: "https://a.example.com"},
{APIKey: "shared-key", BaseURL: "https://b.example.com"},
},
},
configFilePath: writeTestConfigFile(t),
}
rec := httptest.NewRecorder()
c, _ := gin.CreateTestContext(rec)
c.Request = httptest.NewRequest(http.MethodDelete, "/v0/management/vertex-api-key?api-key=shared-key&base-url=https://b.example.com", nil)
h.DeleteVertexCompatKey(c)
if rec.Code != http.StatusOK {
t.Fatalf("status = %d, want %d; body=%s", rec.Code, http.StatusOK, rec.Body.String())
}
if got := len(h.cfg.VertexCompatAPIKey); got != 1 {
t.Fatalf("vertex keys len = %d, want 1", got)
}
if got := h.cfg.VertexCompatAPIKey[0].BaseURL; got != "https://a.example.com" {
t.Fatalf("remaining base-url = %q, want %q", got, "https://a.example.com")
}
}
func TestDeleteCodexKey_RequiresBaseURLWhenAPIKeyDuplicated(t *testing.T) {
t.Parallel()
gin.SetMode(gin.TestMode)
h := &Handler{
cfg: &config.Config{
CodexKey: []config.CodexKey{
{APIKey: "shared-key", BaseURL: "https://a.example.com"},
{APIKey: "shared-key", BaseURL: "https://b.example.com"},
},
},
configFilePath: writeTestConfigFile(t),
}
rec := httptest.NewRecorder()
c, _ := gin.CreateTestContext(rec)
c.Request = httptest.NewRequest(http.MethodDelete, "/v0/management/codex-api-key?api-key=shared-key", nil)
h.DeleteCodexKey(c)
if rec.Code != http.StatusBadRequest {
t.Fatalf("status = %d, want %d; body=%s", rec.Code, http.StatusBadRequest, rec.Body.String())
}
if got := len(h.cfg.CodexKey); got != 2 {
t.Fatalf("codex keys len = %d, want 2", got)
}
}
@@ -12,7 +12,7 @@ import (
"time"
"github.com/gin-gonic/gin"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
// ImportCredential handles uploading a text file with credentials for a specific provider.
+139 -118
View File
@@ -13,12 +13,11 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/buildinfo"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v6/internal/integrations"
"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/internal/buildinfo"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/integrations"
sdkAuth "github.com/router-for-me/CLIProxyAPI/v7/sdk/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
"golang.org/x/crypto/bcrypt"
"gopkg.in/yaml.v3"
)
@@ -44,7 +43,6 @@ type Handler struct {
attemptsMu sync.Mutex
failedAttempts map[string]*attemptInfo // keyed by client IP
authManager *coreauth.Manager
usageStats *usage.RequestStatistics
tokenStore coreauth.Store
localPassword string
allowRemoteOverride bool
@@ -66,7 +64,6 @@ func NewHandler(cfg *config.Config, configFilePath string, manager *coreauth.Man
configFilePath: configFilePath,
failedAttempts: make(map[string]*attemptInfo),
authManager: manager,
usageStats: usage.GetRequestStatistics(),
tokenStore: sdkAuth.GetTokenStore(),
allowRemoteOverride: envSecret != "",
envSecret: envSecret,
@@ -113,6 +110,9 @@ func NewHandlerWithoutConfigFilePath(cfg *config.Config, manager *coreauth.Manag
// SetConfig updates the in-memory config reference when the server hot-reloads.
func (h *Handler) SetConfig(cfg *config.Config) {
clonedCfg := cloneConfig(cfg)
if h == nil {
return
}
h.mu.Lock()
h.cfg = clonedCfg
h.committedCfg = cloneConfig(clonedCfg)
@@ -120,10 +120,14 @@ func (h *Handler) SetConfig(cfg *config.Config) {
}
// SetAuthManager updates the auth manager reference used by management endpoints.
func (h *Handler) SetAuthManager(manager *coreauth.Manager) { h.authManager = manager }
// SetUsageStatistics allows replacing the usage statistics reference.
func (h *Handler) SetUsageStatistics(stats *usage.RequestStatistics) { h.usageStats = stats }
func (h *Handler) SetAuthManager(manager *coreauth.Manager) {
if h == nil {
return
}
h.mu.Lock()
h.authManager = manager
h.mu.Unlock()
}
// SetLocalPassword configures the runtime-local password accepted for localhost requests.
func (h *Handler) SetLocalPassword(password string) { h.localPassword = password }
@@ -150,9 +154,6 @@ func (h *Handler) SetPostAuthHook(hook coreauth.PostAuthHook) {
// All requests (local and remote) require a valid management key.
// Additionally, remote access requires allow-remote-management=true.
func (h *Handler) Middleware() gin.HandlerFunc {
const maxFailures = 5
const banDuration = 30 * time.Minute
return func(c *gin.Context) {
c.Header("X-CPA-VERSION", buildinfo.Version)
c.Header("X-CPA-COMMIT", buildinfo.Commit)
@@ -160,64 +161,6 @@ func (h *Handler) Middleware() gin.HandlerFunc {
clientIP := c.ClientIP()
localClient := clientIP == "127.0.0.1" || clientIP == "::1"
cfg := h.cfg
var (
allowRemote bool
secretHash string
)
if cfg != nil {
allowRemote = cfg.RemoteManagement.AllowRemote
secretHash = cfg.RemoteManagement.SecretKey
}
if h.allowRemoteOverride {
allowRemote = true
}
envSecret := h.envSecret
fail := func() {}
if !localClient {
h.attemptsMu.Lock()
ai := h.failedAttempts[clientIP]
if ai != nil {
if !ai.blockedUntil.IsZero() {
if time.Now().Before(ai.blockedUntil) {
remaining := time.Until(ai.blockedUntil).Round(time.Second)
h.attemptsMu.Unlock()
c.AbortWithStatusJSON(http.StatusForbidden, gin.H{"error": fmt.Sprintf("IP banned due to too many failed attempts. Try again in %s", remaining)})
return
}
// Ban expired, reset state
ai.blockedUntil = time.Time{}
ai.count = 0
}
}
h.attemptsMu.Unlock()
if !allowRemote {
c.AbortWithStatusJSON(http.StatusForbidden, gin.H{"error": "remote management disabled"})
return
}
fail = func() {
h.attemptsMu.Lock()
aip := h.failedAttempts[clientIP]
if aip == nil {
aip = &attemptInfo{}
h.failedAttempts[clientIP] = aip
}
aip.count++
aip.lastActivity = time.Now()
if aip.count >= maxFailures {
aip.blockedUntil = time.Now().Add(banDuration)
aip.count = 0
}
h.attemptsMu.Unlock()
}
}
if secretHash == "" && envSecret == "" {
c.AbortWithStatusJSON(http.StatusForbidden, gin.H{"error": "remote management key not set"})
return
}
// Accept either Authorization: Bearer <key> or X-Management-Key
var provided string
@@ -233,57 +176,116 @@ func (h *Handler) Middleware() gin.HandlerFunc {
provided = c.GetHeader("X-Management-Key")
}
if provided == "" {
if !localClient {
fail()
}
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{"error": "missing management key"})
allowed, statusCode, errMsg := h.AuthenticateManagementKey(clientIP, localClient, provided)
if !allowed {
c.AbortWithStatusJSON(statusCode, gin.H{"error": errMsg})
return
}
if localClient {
if lp := h.localPassword; lp != "" {
if subtle.ConstantTimeCompare([]byte(provided), []byte(lp)) == 1 {
c.Next()
return
}
}
}
if envSecret != "" && subtle.ConstantTimeCompare([]byte(provided), []byte(envSecret)) == 1 {
if !localClient {
h.attemptsMu.Lock()
if ai := h.failedAttempts[clientIP]; ai != nil {
ai.count = 0
ai.blockedUntil = time.Time{}
}
h.attemptsMu.Unlock()
}
c.Next()
return
}
if secretHash == "" || bcrypt.CompareHashAndPassword([]byte(secretHash), []byte(provided)) != nil {
if !localClient {
fail()
}
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{"error": "invalid management key"})
return
}
if !localClient {
h.attemptsMu.Lock()
if ai := h.failedAttempts[clientIP]; ai != nil {
ai.count = 0
ai.blockedUntil = time.Time{}
}
h.attemptsMu.Unlock()
}
c.Next()
}
}
// AuthenticateManagementKey verifies the provided management key for the given client.
// It mirrors the behaviour of Middleware() so non-HTTP callers can reuse the same logic.
func (h *Handler) AuthenticateManagementKey(clientIP string, localClient bool, provided string) (bool, int, string) {
const maxFailures = 5
const banDuration = 30 * time.Minute
if h == nil {
return false, http.StatusForbidden, "remote management disabled"
}
cfg := h.cfg
var (
allowRemote bool
secretHash string
)
if cfg != nil {
allowRemote = cfg.RemoteManagement.AllowRemote
secretHash = cfg.RemoteManagement.SecretKey
}
if h.allowRemoteOverride {
allowRemote = true
}
envSecret := h.envSecret
now := time.Now()
h.attemptsMu.Lock()
ai := h.failedAttempts[clientIP]
if ai != nil && !ai.blockedUntil.IsZero() {
if now.Before(ai.blockedUntil) {
remaining := ai.blockedUntil.Sub(now).Round(time.Second)
h.attemptsMu.Unlock()
return false, http.StatusForbidden, fmt.Sprintf("IP banned due to too many failed attempts. Try again in %s", remaining)
}
// Ban expired, reset state
ai.blockedUntil = time.Time{}
ai.count = 0
}
h.attemptsMu.Unlock()
if !localClient && !allowRemote {
return false, http.StatusForbidden, "remote management disabled"
}
fail := func() {
h.attemptsMu.Lock()
aip := h.failedAttempts[clientIP]
if aip == nil {
aip = &attemptInfo{}
h.failedAttempts[clientIP] = aip
}
aip.count++
aip.lastActivity = time.Now()
if aip.count >= maxFailures {
aip.blockedUntil = time.Now().Add(banDuration)
aip.count = 0
}
h.attemptsMu.Unlock()
}
reset := func() {
h.attemptsMu.Lock()
if ai := h.failedAttempts[clientIP]; ai != nil {
ai.count = 0
ai.blockedUntil = time.Time{}
}
h.attemptsMu.Unlock()
}
if secretHash == "" && envSecret == "" {
return false, http.StatusForbidden, "remote management key not set"
}
if provided == "" {
fail()
return false, http.StatusUnauthorized, "missing management key"
}
if localClient {
if lp := h.localPassword; lp != "" {
if subtle.ConstantTimeCompare([]byte(provided), []byte(lp)) == 1 {
reset()
return true, 0, ""
}
}
}
if envSecret != "" && subtle.ConstantTimeCompare([]byte(provided), []byte(envSecret)) == 1 {
reset()
return true, 0, ""
}
if secretHash == "" || bcrypt.CompareHashAndPassword([]byte(secretHash), []byte(provided)) != nil {
fail()
return false, http.StatusUnauthorized, "invalid management key"
}
reset()
return true, 0, ""
}
// persist saves the current in-memory config to disk.
func (h *Handler) persist(c *gin.Context) bool {
h.mu.Lock()
@@ -311,6 +313,25 @@ func (h *Handler) persistConfig(c *gin.Context, next *config.Config) bool {
return true
}
// persistLocked saves the current in-memory config to disk.
// It expects the caller to hold h.mu.
func (h *Handler) persistLocked(c *gin.Context) bool {
if h.cfg == nil {
c.JSON(http.StatusInternalServerError, gin.H{"error": "config unavailable"})
return false
}
current := cloneConfig(h.cfg)
if err := saveConfigAtomically(h.configFilePath, current); err != nil {
h.cfg = cloneConfig(h.committedCfg)
c.JSON(http.StatusInternalServerError, gin.H{"error": fmt.Sprintf("failed to save config: %v", err)})
return false
}
h.cfg = current
h.committedCfg = cloneConfig(current)
c.JSON(http.StatusOK, gin.H{"status": "ok"})
return true
}
func (h *Handler) mutateConfig(c *gin.Context, mutate func(*config.Config)) bool {
h.mu.Lock()
next := cloneConfig(h.cfg)
@@ -0,0 +1,38 @@
package management
import (
"net/http"
"strings"
"testing"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
func TestAuthenticateManagementKey_LocalhostIPBan_BlocksCorrectKeyDuringBan(t *testing.T) {
h := &Handler{
cfg: &config.Config{},
failedAttempts: make(map[string]*attemptInfo),
envSecret: "test-secret",
}
for i := 0; i < 5; i++ {
allowed, statusCode, errMsg := h.AuthenticateManagementKey("127.0.0.1", true, "wrong-secret")
if allowed {
t.Fatalf("expected auth to be denied at attempt %d", i+1)
}
if statusCode != http.StatusUnauthorized || errMsg != "invalid management key" {
t.Fatalf("unexpected auth failure at attempt %d: status=%d msg=%q", i+1, statusCode, errMsg)
}
}
allowed, statusCode, errMsg := h.AuthenticateManagementKey("127.0.0.1", true, "test-secret")
if allowed {
t.Fatalf("expected correct key to be denied while banned")
}
if statusCode != http.StatusForbidden {
t.Fatalf("expected forbidden status while banned, got %d", statusCode)
}
if !strings.HasPrefix(errMsg, "IP banned due to too many failed attempts. Try again in") {
t.Fatalf("unexpected banned message: %q", errMsg)
}
}
+1 -1
View File
@@ -5,7 +5,7 @@ import (
"strings"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
)
// GetHarnessFiles returns the list of harness files for a session.
@@ -4,7 +4,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/integrations"
"github.com/router-for-me/CLIProxyAPI/v7/internal/integrations"
)
// GetIntegrationsStatus returns the list of detected tools and their configuration status.
+1 -1
View File
@@ -13,7 +13,7 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
)
const (
@@ -12,7 +12,7 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
func TestDeleteMemorySession_RejectsTraversal(t *testing.T) {
+1 -1
View File
@@ -12,7 +12,7 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
)
func (h *Handler) ListMemorySessions(c *gin.Context) {
@@ -6,7 +6,7 @@ import (
"strconv"
"strings"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
)
func memoryBaseDir() string {
@@ -5,7 +5,7 @@ import (
"strings"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v7/internal/registry"
)
// GetStaticModelDefinitions returns static model metadata for a given channel.
@@ -4,7 +4,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
// ModelMappingEntry represents a global model mapping for the API
@@ -225,12 +225,8 @@ func NormalizeOAuthProvider(provider string) (string, error) {
return "codex", nil
case "gemini", "google":
return "gemini", nil
case "iflow", "i-flow":
return "iflow", nil
case "antigravity", "anti-gravity":
return "antigravity", nil
case "qwen":
return "qwen", nil
default:
return "", errUnsupportedOAuthFlow
}
@@ -11,7 +11,7 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/buildinfo"
"github.com/router-for-me/CLIProxyAPI/v7/internal/buildinfo"
)
// GetProxyPilotLogTail returns tail lines for the launcher-managed stdout/stderr logs.
+1 -1
View File
@@ -2,7 +2,7 @@ package management
import (
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
// Quota exceeded toggles
@@ -7,7 +7,7 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/api/middleware"
"github.com/router-for-me/CLIProxyAPI/v7/internal/api/middleware"
)
// GetRequestHistory returns the persistent request history with optional filtering.
+1 -1
View File
@@ -4,7 +4,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/api/middleware"
"github.com/router-for-me/CLIProxyAPI/v7/internal/api/middleware"
)
// GetRequests returns the list of recent requests from the in-memory ring buffer.
@@ -10,7 +10,7 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
)
func (h *Handler) GetSemanticHealth(c *gin.Context) {
@@ -4,7 +4,7 @@ import (
"context"
"sync"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
type memoryAuthStore struct {
@@ -4,7 +4,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
)
// ThinkingBudgetResponse represents the thinking budget settings response.
+1 -1
View File
@@ -5,7 +5,7 @@ import (
"sync"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/updates"
"github.com/router-for-me/CLIProxyAPI/v7/internal/updates"
)
// updateState tracks the current update operation state.
+33 -57
View File
@@ -2,78 +2,54 @@ package management
import (
"encoding/json"
"errors"
"net/http"
"time"
"strconv"
"strings"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
"github.com/router-for-me/CLIProxyAPI/v7/internal/redisqueue"
)
type usageExportPayload struct {
Version int `json:"version"`
ExportedAt time.Time `json:"exported_at"`
Usage usage.StatisticsSnapshot `json:"usage"`
}
type usageQueueRecord []byte
type usageImportPayload struct {
Version int `json:"version"`
Usage usage.StatisticsSnapshot `json:"usage"`
}
// GetUsageStatistics returns the in-memory request statistics snapshot.
func (h *Handler) GetUsageStatistics(c *gin.Context) {
var snapshot usage.StatisticsSnapshot
if h != nil && h.usageStats != nil {
snapshot = h.usageStats.Snapshot()
func (r usageQueueRecord) MarshalJSON() ([]byte, error) {
if json.Valid(r) {
return append([]byte(nil), r...), nil
}
c.JSON(http.StatusOK, gin.H{
"usage": snapshot,
"failed_requests": snapshot.FailureCount,
})
return json.Marshal(string(r))
}
// ExportUsageStatistics returns a complete usage snapshot for backup/migration.
func (h *Handler) ExportUsageStatistics(c *gin.Context) {
var snapshot usage.StatisticsSnapshot
if h != nil && h.usageStats != nil {
snapshot = h.usageStats.Snapshot()
}
c.JSON(http.StatusOK, usageExportPayload{
Version: 1,
ExportedAt: time.Now().UTC(),
Usage: snapshot,
})
}
// ImportUsageStatistics merges a previously exported usage snapshot into memory.
func (h *Handler) ImportUsageStatistics(c *gin.Context) {
if h == nil || h.usageStats == nil {
c.JSON(http.StatusBadRequest, gin.H{"error": "usage statistics unavailable"})
// GetUsageQueue pops queued usage records from the usage queue.
func (h *Handler) GetUsageQueue(c *gin.Context) {
if h == nil {
c.JSON(http.StatusInternalServerError, gin.H{"error": "handler unavailable"})
return
}
data, err := c.GetRawData()
if err != nil {
c.JSON(http.StatusBadRequest, gin.H{"error": "failed to read request body"})
count, errCount := parseUsageQueueCount(c.Query("count"))
if errCount != nil {
c.JSON(http.StatusBadRequest, gin.H{"error": errCount.Error()})
return
}
var payload usageImportPayload
if err := json.Unmarshal(data, &payload); err != nil {
c.JSON(http.StatusBadRequest, gin.H{"error": "invalid json"})
return
}
if payload.Version != 0 && payload.Version != 1 {
c.JSON(http.StatusBadRequest, gin.H{"error": "unsupported version"})
return
items := redisqueue.PopOldest(count)
records := make([]usageQueueRecord, 0, len(items))
for _, item := range items {
records = append(records, usageQueueRecord(append([]byte(nil), item...)))
}
result := h.usageStats.MergeSnapshot(payload.Usage)
snapshot := h.usageStats.Snapshot()
c.JSON(http.StatusOK, gin.H{
"added": result.Added,
"skipped": result.Skipped,
"total_requests": snapshot.TotalRequests,
"failed_requests": snapshot.FailureCount,
})
c.JSON(http.StatusOK, records)
}
func parseUsageQueueCount(value string) (int, error) {
value = strings.TrimSpace(value)
if value == "" {
return 1, nil
}
count, errCount := strconv.Atoi(value)
if errCount != nil || count <= 0 {
return 0, errors.New("count must be a positive integer")
}
return count, nil
}
@@ -0,0 +1,98 @@
package management
import (
"encoding/json"
"net/http"
"net/http/httptest"
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v7/internal/redisqueue"
)
func TestGetUsageQueuePopsRequestedRecords(t *testing.T) {
gin.SetMode(gin.TestMode)
withManagementUsageQueue(t, func() {
redisqueue.Enqueue([]byte(`{"id":1}`))
redisqueue.Enqueue([]byte(`{"id":2}`))
redisqueue.Enqueue([]byte(`{"id":3}`))
rec := httptest.NewRecorder()
ginCtx, _ := gin.CreateTestContext(rec)
ginCtx.Request = httptest.NewRequest(http.MethodGet, "/v0/management/usage-queue?count=2", nil)
h := &Handler{}
h.GetUsageQueue(ginCtx)
if rec.Code != http.StatusOK {
t.Fatalf("status = %d, want %d body=%s", rec.Code, http.StatusOK, rec.Body.String())
}
var payload []json.RawMessage
if errUnmarshal := json.Unmarshal(rec.Body.Bytes(), &payload); errUnmarshal != nil {
t.Fatalf("unmarshal response: %v", errUnmarshal)
}
if len(payload) != 2 {
t.Fatalf("response records = %d, want 2", len(payload))
}
requireRecordID(t, payload[0], 1)
requireRecordID(t, payload[1], 2)
remaining := redisqueue.PopOldest(10)
if len(remaining) != 1 || string(remaining[0]) != `{"id":3}` {
t.Fatalf("remaining queue = %q, want third item only", remaining)
}
})
}
func TestGetUsageQueueInvalidCountDoesNotPop(t *testing.T) {
gin.SetMode(gin.TestMode)
withManagementUsageQueue(t, func() {
redisqueue.Enqueue([]byte(`{"id":1}`))
rec := httptest.NewRecorder()
ginCtx, _ := gin.CreateTestContext(rec)
ginCtx.Request = httptest.NewRequest(http.MethodGet, "/v0/management/usage-queue?count=0", nil)
h := &Handler{}
h.GetUsageQueue(ginCtx)
if rec.Code != http.StatusBadRequest {
t.Fatalf("status = %d, want %d body=%s", rec.Code, http.StatusBadRequest, rec.Body.String())
}
remaining := redisqueue.PopOldest(10)
if len(remaining) != 1 || string(remaining[0]) != `{"id":1}` {
t.Fatalf("remaining queue = %q, want original item", remaining)
}
})
}
func withManagementUsageQueue(t *testing.T, fn func()) {
t.Helper()
prevQueueEnabled := redisqueue.Enabled()
redisqueue.SetEnabled(false)
redisqueue.SetEnabled(true)
defer func() {
redisqueue.SetEnabled(false)
redisqueue.SetEnabled(prevQueueEnabled)
}()
fn()
}
func requireRecordID(t *testing.T, raw json.RawMessage, want int) {
t.Helper()
var payload struct {
ID int `json:"id"`
}
if errUnmarshal := json.Unmarshal(raw, &payload); errUnmarshal != nil {
t.Fatalf("unmarshal record: %v", errUnmarshal)
}
if payload.ID != want {
t.Fatalf("record id = %d, want %d", payload.ID, want)
}
}
@@ -9,8 +9,8 @@ import (
"strings"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/vertex"
coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
"github.com/router-for-me/CLIProxyAPI/v7/internal/auth/vertex"
coreauth "github.com/router-for-me/CLIProxyAPI/v7/sdk/cliproxy/auth"
)
// ImportVertexCredential handles uploading a Vertex service account JSON and saving it as an auth record.
+1 -1
View File
@@ -5,7 +5,7 @@ import (
"net/http"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/translator"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/translator"
)
// TranslatorHandler provides HTTP handlers for translation-related endpoints.
@@ -8,7 +8,7 @@ import (
"testing"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/translator"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/translator"
)
func init() {
@@ -4,9 +4,9 @@ import (
"strconv"
"strings"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v6/internal/runtime/executor"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v7/internal/runtime/executor"
"github.com/tidwall/gjson"
"github.com/tidwall/sjson"
)
@@ -9,10 +9,10 @@ import (
"sync"
"time"
"github.com/router-for-me/CLIProxyAPI/v6/internal/embeddings"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
"github.com/router-for-me/CLIProxyAPI/v7/internal/embeddings"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/registry"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
)
const (
@@ -14,7 +14,7 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/memory"
"github.com/router-for-me/CLIProxyAPI/v7/internal/memory"
"github.com/tidwall/gjson"
"github.com/tidwall/sjson"
)
+2 -2
View File
@@ -11,8 +11,8 @@ import (
"sync/atomic"
"time"
"github.com/router-for-me/CLIProxyAPI/v6/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
"github.com/router-for-me/CLIProxyAPI/v7/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v7/internal/usage"
)
const (
+2 -2
View File
@@ -11,8 +11,8 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
)
const maxErrorOnlyCapturedRequestBodyBytes int64 = 1 << 20 // 1 MiB
+64 -18
View File
@@ -10,11 +10,13 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
"github.com/router-for-me/CLIProxyAPI/v7/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
)
const requestBodyOverrideContextKey = "REQUEST_BODY_OVERRIDE"
const responseBodyOverrideContextKey = "RESPONSE_BODY_OVERRIDE"
const websocketTimelineOverrideContextKey = "WEBSOCKET_TIMELINE_OVERRIDE"
// RequestInfo holds essential details of an incoming HTTP request for logging purposes.
type RequestInfo struct {
@@ -304,6 +306,10 @@ func (w *ResponseWriterWrapper) Finalize(c *gin.Context) error {
if len(apiResponse) > 0 {
_ = w.streamWriter.WriteAPIResponse(apiResponse)
}
apiWebsocketTimeline := w.extractAPIWebsocketTimeline(c)
if len(apiWebsocketTimeline) > 0 {
_ = w.streamWriter.WriteAPIWebsocketTimeline(apiWebsocketTimeline)
}
if err := w.streamWriter.Close(); err != nil {
w.streamWriter = nil
return err
@@ -312,7 +318,7 @@ func (w *ResponseWriterWrapper) Finalize(c *gin.Context) error {
return nil
}
return w.logRequest(w.extractRequestBody(c), finalStatusCode, w.cloneHeaders(), w.body.Bytes(), w.extractAPIRequest(c), w.extractAPIResponse(c), w.extractAPIResponseTimestamp(c), slicesAPIResponseError, forceLog)
return w.logRequest(w.extractRequestBody(c), finalStatusCode, w.cloneHeaders(), w.extractResponseBody(c), w.extractWebsocketTimeline(c), w.extractAPIRequest(c), w.extractAPIResponse(c), w.extractAPIWebsocketTimeline(c), w.extractAPIResponseTimestamp(c), slicesAPIResponseError, forceLog)
}
func (w *ResponseWriterWrapper) cloneHeaders() map[string][]string {
@@ -352,6 +358,18 @@ func (w *ResponseWriterWrapper) extractAPIResponse(c *gin.Context) []byte {
return data
}
func (w *ResponseWriterWrapper) extractAPIWebsocketTimeline(c *gin.Context) []byte {
apiTimeline, isExist := c.Get("API_WEBSOCKET_TIMELINE")
if !isExist {
return nil
}
data, ok := apiTimeline.([]byte)
if !ok || len(data) == 0 {
return nil
}
return bytes.Clone(data)
}
func (w *ResponseWriterWrapper) extractAPIResponseTimestamp(c *gin.Context) time.Time {
ts, isExist := c.Get("API_RESPONSE_TIMESTAMP")
if !isExist {
@@ -364,19 +382,8 @@ func (w *ResponseWriterWrapper) extractAPIResponseTimestamp(c *gin.Context) time
}
func (w *ResponseWriterWrapper) extractRequestBody(c *gin.Context) []byte {
if c != nil {
if bodyOverride, isExist := c.Get(requestBodyOverrideContextKey); isExist {
switch value := bodyOverride.(type) {
case []byte:
if len(value) > 0 {
return bytes.Clone(value)
}
case string:
if strings.TrimSpace(value) != "" {
return []byte(value)
}
}
}
if body := extractBodyOverride(c, requestBodyOverrideContextKey); len(body) > 0 {
return body
}
if w.requestInfo != nil && len(w.requestInfo.Body) > 0 {
return w.requestInfo.Body
@@ -384,13 +391,48 @@ func (w *ResponseWriterWrapper) extractRequestBody(c *gin.Context) []byte {
return nil
}
func (w *ResponseWriterWrapper) logRequest(requestBody []byte, statusCode int, headers map[string][]string, body []byte, apiRequestBody, apiResponseBody []byte, apiResponseTimestamp time.Time, apiResponseErrors []*interfaces.ErrorMessage, forceLog bool) error {
func (w *ResponseWriterWrapper) extractResponseBody(c *gin.Context) []byte {
if body := extractBodyOverride(c, responseBodyOverrideContextKey); len(body) > 0 {
return body
}
if w.body == nil || w.body.Len() == 0 {
return nil
}
return bytes.Clone(w.body.Bytes())
}
func (w *ResponseWriterWrapper) extractWebsocketTimeline(c *gin.Context) []byte {
return extractBodyOverride(c, websocketTimelineOverrideContextKey)
}
func extractBodyOverride(c *gin.Context, key string) []byte {
if c == nil {
return nil
}
bodyOverride, isExist := c.Get(key)
if !isExist {
return nil
}
switch value := bodyOverride.(type) {
case []byte:
if len(value) > 0 {
return bytes.Clone(value)
}
case string:
if strings.TrimSpace(value) != "" {
return []byte(value)
}
}
return nil
}
func (w *ResponseWriterWrapper) logRequest(requestBody []byte, statusCode int, headers map[string][]string, body, websocketTimeline, apiRequestBody, apiResponseBody, apiWebsocketTimeline []byte, apiResponseTimestamp time.Time, apiResponseErrors []*interfaces.ErrorMessage, forceLog bool) error {
if w.requestInfo == nil {
return nil
}
if loggerWithOptions, ok := w.logger.(interface {
LogRequestWithOptions(string, string, map[string][]string, []byte, int, map[string][]string, []byte, []byte, []byte, []*interfaces.ErrorMessage, bool, string, time.Time, time.Time) error
LogRequestWithOptions(string, string, map[string][]string, []byte, int, map[string][]string, []byte, []byte, []byte, []byte, []byte, []*interfaces.ErrorMessage, bool, string, time.Time, time.Time) error
}); ok {
return loggerWithOptions.LogRequestWithOptions(
w.requestInfo.URL,
@@ -400,8 +442,10 @@ func (w *ResponseWriterWrapper) logRequest(requestBody []byte, statusCode int, h
statusCode,
headers,
body,
websocketTimeline,
apiRequestBody,
apiResponseBody,
apiWebsocketTimeline,
apiResponseErrors,
forceLog,
w.requestInfo.RequestID,
@@ -418,8 +462,10 @@ func (w *ResponseWriterWrapper) logRequest(requestBody []byte, statusCode int, h
statusCode,
headers,
body,
websocketTimeline,
apiRequestBody,
apiResponseBody,
apiWebsocketTimeline,
apiResponseErrors,
w.requestInfo.RequestID,
w.requestInfo.Timestamp,
+160 -1
View File
@@ -1,10 +1,14 @@
package middleware
import (
"bytes"
"net/http/httptest"
"testing"
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v7/internal/interfaces"
"github.com/router-for-me/CLIProxyAPI/v7/internal/logging"
)
func TestExtractRequestBodyPrefersOverride(t *testing.T) {
@@ -33,7 +37,7 @@ func TestExtractRequestBodySupportsStringOverride(t *testing.T) {
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
wrapper := &ResponseWriterWrapper{}
wrapper := &ResponseWriterWrapper{body: &bytes.Buffer{}}
c.Set(requestBodyOverrideContextKey, "override-as-string")
body := wrapper.extractRequestBody(c)
@@ -41,3 +45,158 @@ func TestExtractRequestBodySupportsStringOverride(t *testing.T) {
t.Fatalf("request body = %q, want %q", string(body), "override-as-string")
}
}
func TestExtractResponseBodyPrefersOverride(t *testing.T) {
gin.SetMode(gin.TestMode)
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
wrapper := &ResponseWriterWrapper{body: &bytes.Buffer{}}
wrapper.body.WriteString("original-response")
body := wrapper.extractResponseBody(c)
if string(body) != "original-response" {
t.Fatalf("response body = %q, want %q", string(body), "original-response")
}
c.Set(responseBodyOverrideContextKey, []byte("override-response"))
body = wrapper.extractResponseBody(c)
if string(body) != "override-response" {
t.Fatalf("response body = %q, want %q", string(body), "override-response")
}
body[0] = 'X'
if got := wrapper.extractResponseBody(c); string(got) != "override-response" {
t.Fatalf("response override should be cloned, got %q", string(got))
}
}
func TestExtractResponseBodySupportsStringOverride(t *testing.T) {
gin.SetMode(gin.TestMode)
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
wrapper := &ResponseWriterWrapper{}
c.Set(responseBodyOverrideContextKey, "override-response-as-string")
body := wrapper.extractResponseBody(c)
if string(body) != "override-response-as-string" {
t.Fatalf("response body = %q, want %q", string(body), "override-response-as-string")
}
}
func TestExtractBodyOverrideClonesBytes(t *testing.T) {
gin.SetMode(gin.TestMode)
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
override := []byte("body-override")
c.Set(requestBodyOverrideContextKey, override)
body := extractBodyOverride(c, requestBodyOverrideContextKey)
if !bytes.Equal(body, override) {
t.Fatalf("body override = %q, want %q", string(body), string(override))
}
body[0] = 'X'
if !bytes.Equal(override, []byte("body-override")) {
t.Fatalf("override mutated: %q", string(override))
}
}
func TestExtractWebsocketTimelineUsesOverride(t *testing.T) {
gin.SetMode(gin.TestMode)
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
wrapper := &ResponseWriterWrapper{}
if got := wrapper.extractWebsocketTimeline(c); got != nil {
t.Fatalf("expected nil websocket timeline, got %q", string(got))
}
c.Set(websocketTimelineOverrideContextKey, []byte("timeline"))
body := wrapper.extractWebsocketTimeline(c)
if string(body) != "timeline" {
t.Fatalf("websocket timeline = %q, want %q", string(body), "timeline")
}
}
func TestFinalizeStreamingWritesAPIWebsocketTimeline(t *testing.T) {
gin.SetMode(gin.TestMode)
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
streamWriter := &testStreamingLogWriter{}
wrapper := &ResponseWriterWrapper{
ResponseWriter: c.Writer,
logger: &testRequestLogger{enabled: true},
requestInfo: &RequestInfo{
URL: "/v1/responses",
Method: "POST",
Headers: map[string][]string{"Content-Type": {"application/json"}},
RequestID: "req-1",
Timestamp: time.Date(2026, time.April, 1, 12, 0, 0, 0, time.UTC),
},
isStreaming: true,
streamWriter: streamWriter,
}
c.Set("API_WEBSOCKET_TIMELINE", []byte("Timestamp: 2026-04-01T12:00:00Z\nEvent: api.websocket.request\n{}"))
if err := wrapper.Finalize(c); err != nil {
t.Fatalf("Finalize error: %v", err)
}
if string(streamWriter.apiWebsocketTimeline) != "Timestamp: 2026-04-01T12:00:00Z\nEvent: api.websocket.request\n{}" {
t.Fatalf("stream writer websocket timeline = %q", string(streamWriter.apiWebsocketTimeline))
}
if !streamWriter.closed {
t.Fatal("expected stream writer to be closed")
}
}
type testRequestLogger struct {
enabled bool
}
func (l *testRequestLogger) LogRequest(string, string, map[string][]string, []byte, int, map[string][]string, []byte, []byte, []byte, []byte, []byte, []*interfaces.ErrorMessage, string, time.Time, time.Time) error {
return nil
}
func (l *testRequestLogger) LogStreamingRequest(string, string, map[string][]string, []byte, string) (logging.StreamingLogWriter, error) {
return &testStreamingLogWriter{}, nil
}
func (l *testRequestLogger) IsEnabled() bool {
return l.enabled
}
type testStreamingLogWriter struct {
apiWebsocketTimeline []byte
closed bool
}
func (w *testStreamingLogWriter) WriteChunkAsync([]byte) {}
func (w *testStreamingLogWriter) WriteStatus(int, map[string][]string) error {
return nil
}
func (w *testStreamingLogWriter) WriteAPIRequest([]byte) error {
return nil
}
func (w *testStreamingLogWriter) WriteAPIResponse([]byte) error {
return nil
}
func (w *testStreamingLogWriter) WriteAPIWebsocketTimeline(apiWebsocketTimeline []byte) error {
w.apiWebsocketTimeline = bytes.Clone(apiWebsocketTimeline)
return nil
}
func (w *testStreamingLogWriter) SetFirstChunkTimestamp(time.Time) {}
func (w *testStreamingLogWriter) Close() error {
w.closed = true
return nil
}
+3 -3
View File
@@ -9,9 +9,9 @@ import (
"sync"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/api/modules"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v6/sdk/access"
"github.com/router-for-me/CLIProxyAPI/v7/internal/api/modules"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v7/sdk/access"
log "github.com/sirupsen/logrus"
)
+4 -4
View File
@@ -9,10 +9,10 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/api/modules"
"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v6/sdk/access"
"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers"
"github.com/router-for-me/CLIProxyAPI/v7/internal/api/modules"
"github.com/router-for-me/CLIProxyAPI/v7/internal/config"
sdkaccess "github.com/router-for-me/CLIProxyAPI/v7/sdk/access"
"github.com/router-for-me/CLIProxyAPI/v7/sdk/api/handlers"
)
func TestAmpModule_Name(t *testing.T) {
+14 -2
View File
@@ -8,8 +8,8 @@ import (
"time"
"github.com/gin-gonic/gin"
"github.com/router-for-me/CLIProxyAPI/v6/internal/thinking"
"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
"github.com/router-for-me/CLIProxyAPI/v7/internal/thinking"
"github.com/router-for-me/CLIProxyAPI/v7/internal/util"
log "github.com/sirupsen/logrus"
"github.com/tidwall/gjson"
"github.com/tidwall/sjson"
@@ -123,6 +123,10 @@ func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc
return
}
// Sanitize request body: remove thinking blocks with invalid signatures
// to prevent upstream API 400 errors
bodyBytes = SanitizeAmpRequestBody(bodyBytes)
// Restore the body for the handler to read
c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))
@@ -249,6 +253,7 @@ func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc
log.Debugf("amp model mapping: request %s -> %s", normalizedModel, resolvedModel)
logAmpRouting(RouteTypeModelMapping, modelName, resolvedModel, providerName, requestPath)
rewriter := NewResponseRewriter(c.Writer, modelName)
rewriter.suppressThinking = true
c.Writer = rewriter
// Filter Anthropic-Beta header only for local handling paths
filterAntropicBetaHeader(c)
@@ -259,10 +264,17 @@ func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc
} else if len(providers) > 0 {
// Log: Using local provider (free)
logAmpRouting(RouteTypeLocalProvider, modelName, resolvedModel, providerName, requestPath)
// Wrap with ResponseRewriter for local providers too, because upstream
// proxies (e.g. NewAPI) may return a different model name and lack
// Amp-required fields like thinking.signature.
rewriter := NewResponseRewriter(c.Writer, modelName)
rewriter.suppressThinking = providerName != "claude"
c.Writer = rewriter
// Filter Anthropic-Beta header only for local handling paths
filterAntropicBetaHeader(c)
c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))
handler(c)
rewriter.Flush()
} else {
// No provider, no mapping, no proxy: fall back to the wrapped handler so it can return an error response
c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))

Some files were not shown because too many files have changed in this diff Show More