Stop juggling API keys. ProxyPilot is a powerful local API proxy that lets you use your existing Claude Code, Codex, Gemini, Kiro, and Qwen subscriptions with any AI coding tool โ no separate API keys required.
Built in Go, it handles OAuth authentication, token management, and API translation automatically. One server to route them all.
--- > [!TIP] > ๐ฃ **Latest models supported:** > Claude Opus 4.5 / Sonnet 4.5 with extended thinking, GPT-5.2 / GPT-5.2 Codex, Gemini 3 Pro/Flash, and Kiro (AWS CodeWhisperer)! ๐ **Setup Guides:** - [Claude Code Setup โ](docs/claude-code-local-proxy.md) - [Cursor IDE Setup โ](docs/cursor-ide.md) --- ## Features - ๐ฏ **10 Auth Providers** - Claude, Codex (OpenAI), Gemini, Gemini CLI, Kiro (AWS), Amazon Q CLI, Qwen, Antigravity, MiniMax, Zhipu AI - ๐ **Universal API Translation** - Auto-converts between OpenAI, Anthropic, and Gemini formats - ๐ง **Tool Calling Repair** - Fixes tool/function call mismatches between providers automatically - ๐ง **Extended Thinking** - Full support for Claude and Gemini thinking models - ๐ **OAuth Integration** - Browser-based login with automatic token refresh - ๐ฅ **Multi-Account Support** - Round-robin distribution with automatic failover - โก **Quota Auto-Switch** - Automatically switches to backup project/model when quota exceeded - ๐ **Usage Statistics** - Track requests, tokens, and errors per provider/model - ๐งฉ **Context Compression** - LLM-based summarization for long sessions (Factory.ai research) - ๐ค **Agentic Harness** - Guided workflow for coding agents (Anthropic research) - ๐พ **Session Memory** - Persistent storage across conversation turns - ๐จ **System Tray** - Native Windows tray app for quick access - ๐ก **60+ Management APIs** - Full control via REST endpoints --- ## Supported Providers | Provider | Auth Method | Models | |----------|-------------|--------| | Claude (Anthropic) | OAuth2 / API Key | Claude Opus 4.5, Sonnet 4.5, Haiku 4.5 | | Codex (OpenAI) | OAuth2 / API Key | GPT-5.2, GPT-5.2 Codex | | Gemini | OAuth2 / API Key | Gemini 3 Pro, Gemini 3 Flash | | Gemini CLI | OAuth2 | Cloud Code Assist (separate quota) | | Kiro | OAuth2 + AWS SSO | AWS CodeWhisperer | | Amazon Q CLI | Import from CLI | Amazon Q Developer | | Qwen | OAuth2 | Qwen models | | Antigravity | OAuth2 | Gemini via Antigravity (separate quota) | | MiniMax | API Key | MiniMax M2, M2.1 models | | Zhipu AI | API Key | GLM-4.5, GLM-4.6, GLM-4.7 | | Custom | API Key | Any OpenAI-compatible endpoint | --- ## Installation ### Download Pre-built Release (Recommended) 1. Go to the [**Releases**](https://github.com/Finesssee/ProxyPilot/releases) page 2. Download the latest binary for your platform 3. Run `./proxypilot` ### Build from Source ```bash git clone https://github.com/Finesssee/ProxyPilot.git cd ProxyPilot go build -o proxypilot ./cmd/server ./proxypilot ``` --- ## Usage ### 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/proxypilot.html` ### Authentication Run OAuth login for your provider: ```bash # OAuth providers (opens browser) ./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 --qwen-login # Qwen ./proxypilot --antigravity-login # Antigravity # Import providers (from existing CLI tools) ./proxypilot --amazonq-import # Amazon Q CLI (run 'q login' first) ./proxypilot --kiro-import # Kiro IDE token # API key providers (prompts for key) ./proxypilot --minimax-login # MiniMax API key ./proxypilot --zhipu-login # Zhipu AI API key ``` OAuth tokens are stored locally and auto-refreshed before expiry. ### Security Defaults (Auth + CORS) - Proxy requests require API keys by default. To allow unauthenticated access (not recommended), set `allow-unauthenticated: true` in `config.yaml`. - CORS is enabled for non-management endpoints by default (wildcard `*`). Management endpoints **do not** emit CORS headers unless you explicitly allow origins under `cors.management-allow-origins`. Example: ```yaml allow-unauthenticated: false cors: allow-origins: - "http://localhost:5173" management-allow-origins: - "http://localhost:5173" ``` ### Configure Your Tools **Claude Code** (`~/.claude/settings.json`): ```json { "env": { "ANTHROPIC_BASE_URL": "http://127.0.0.1:8317", "ANTHROPIC_AUTH_TOKEN": "your-api-key" } } ``` **Codex CLI** (`~/.codex/config.toml`): ```toml [openai] api_base_url = "http://127.0.0.1:8317" ``` **Factory Droid** (`~/.factory/settings.json`): ```json { "customModels": [{ "name": "ProxyPilot", "baseUrl": "http://127.0.0.1:8317" }] } ``` --- ## API Endpoints ``` POST /v1/chat/completions # OpenAI Chat Completions POST /v1/responses # OpenAI Responses API POST /v1/messages # Anthropic Messages API GET /v1/models # List available models GET /healthz # Health check ``` All endpoints auto-translate between formats based on the target provider. --- ## Caching ProxyPilot includes two caching layers to reduce latency and token usage. ### Response Cache Caches full API responses for identical requests. Useful for repeated queries during development. **Config** (`config.yaml`): ```yaml response-cache: enabled: true # Default: false max-size: 1000 # Max entries (default: 1000) max-bytes: 0 # Optional total cache size cap in bytes ttl-seconds: 300 # Cache TTL (default: 300 = 5 min) exclude-models: # Models to skip (supports wildcards) - "*-thinking" - "o1-*" ``` ### Prompt Cache Synthetic prompt caching for providers without native support. Tracks repeated system prompts and estimates token savings. **Config** (`config.yaml`): ```yaml prompt-cache: enabled: true # Default: false max-size: 500 # Max entries (default: 500) max-bytes: 0 # Optional total cache size cap in bytes ttl-seconds: 1800 # Cache TTL (default: 1800 = 30 min) ``` ### Cache Management API | Endpoint | Method | Description | |----------|--------|-------------| | `/v0/management/cache/stats` | GET | Response cache stats (hits, misses, size) | | `/v0/management/cache/clear` | POST | Clear response cache | | `/v0/management/cache/enabled` | PUT | Enable/disable at runtime `{"enabled": true}` | | `/v0/management/prompt-cache/stats` | GET | Prompt cache stats + estimated tokens saved | | `/v0/management/prompt-cache/clear` | POST | Clear prompt cache | | `/v0/management/prompt-cache/enabled` | PUT | Enable/disable at runtime | | `/v0/management/prompt-cache/top` | GET | Top 10 most-hit prompts | --- ## Lightweight Profile For lowโmemory or highโthroughput setups, you can disable heavier features: ```yaml commercial-mode: true usage-statistics-enabled: false usage-sample-rate: 0.25 metrics-enabled: false request-history-enabled: false request-history-sample-rate: 0.25 agentic-harness-enabled: false prompt-budget-enabled: false request-log: false response-cache: enabled: false prompt-cache: enabled: false ``` --- ## CLI Tools | Binary | Description | |--------|-------------| | `proxypilot` | Main CLI with server and config switching | | `proxypilot-tray` | System tray app | --- ## Tool Integrations Works with these AI coding tools: - **Claude Code** - Auto-configure via settings.json - **Codex CLI** - Auto-configure via config.toml - **Factory Droid** - Auto-configure via settings.json - **Cursor IDE** - Manual endpoint configuration - **Continue** - Manual endpoint configuration --- ## Requirements - macOS, Linux, or Windows - Go 1.24+ (for building from source) --- ## Author ProxyPilot is developed and maintained by [@Finesssee](https://github.com/Finesssee). Some contributors shown in the git history are from upstream [CLIProxyAPI](https://github.com/router-for-me/CLIProxyAPI) merges. Direct contributors to ProxyPilot will be listed here as the project grows. --- ## Credits ProxyPilot builds upon excellent work from the open-source community: - **[CLIProxyAPI](https://github.com/router-for-me/CLIProxyAPI)** - The original unified proxy server that inspired this project - **[VibeProxy](https://github.com/automazeio/vibeproxy)** - Native macOS menu bar app showcasing clean proxy UX Long-context features are inspired by research from the AI community: - **[Factory.ai](https://factory.ai/news/evaluating-compression)** - Context compression techniques for long-running coding agents - **[Anthropic](https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents)** - Effective harnesses for long-running agents Special thanks to these teams for sharing their work and insights. --- ## License MIT License - see [LICENSE](LICENSE) for details. --- ## Support - **Report Issues**: [GitHub Issues](https://github.com/Finesssee/ProxyPilot/issues) --- ## Disclaimer This project is for educational and interoperability research purposes. It interacts with various APIs to provide compatibility layers. - **Use at your own risk.** Authors are not responsible for account suspensions or service interruptions. - **Not affiliated** with Google, OpenAI, Anthropic, Amazon, or any other provider. - Users must comply with the Terms of Service of connected platforms.