Replace silent execFileSync with async spawn + --progress flag so the
spinner shows real download progress (e.g. "Receiving objects 45%").
For large monorepos, try partial clone (--filter=blob:none) with sparse
checkout of SKILL_DISCOVERY_PATHS first, falling back to regular
--depth 1 clone if the server doesn't support it. Also adds
--single-branch --no-tags to all clones.
Shared cloneRepo() helper in base.ts eliminates duplicate clone logic
across all four git providers.
- selectSkills/selectAgents now return { result, exitCode } to
distinguish errors (exit 1) from cancellations (exit 0)
- Fixes: --skill with nonexistent name returned 0 instead of 1
- Fixes: --agent with invalid name returned 0 instead of 1
- ai.ts: s.stop('msg', 1) → s.stop(colors.error('msg'))
(clack spinner.stop only accepts message, not exit code)
generate.ts and manifest.ts now warn when --agents (deprecated) is
used instead of --agent. Message: "Warning: --agents is deprecated,
use --agent instead (--agents will be removed in v2.0)"
Findings 3 (scan.ts stderr) and 4 (translate.ts stderr) skipped:
error() from onboarding goes to stdout via clack — this is the
standard pattern across all migrated commands. Changing individual
commands would break consistency.
* fix: version display, install UX — audit findings
- Fix version showing hardcoded 1.7.6 instead of actual 1.19.2
- Export setVersion/setAgentCount from @skillkit/cli
- Call setVersion(version) and setAgentCount(getAllAdapters().length) in cli.ts
- Change fallback from '1.7.6' to 'dev' so stale values are obvious
- Fix install pre-selecting all skills by default (bad UX)
- Add "Install all / Select specific" prompt before multiselect
- When selecting manually, start with nothing pre-selected
- Apply same fix to both InstallCommand and runInstallFlow()
- Matches existing quickAgentSelect pattern
Fixes found during comprehensive codebase audit.
Related issues: #87, #88, #89, #90, #91, #92, #93, #94
* refactor: extract quickSkillSelect, use getAdapterCount
- Extract duplicated "Install all / Select specific" pattern into
quickSkillSelect() in prompts.ts (mirrors existing quickAgentSelect)
- Both install.ts and runInstallFlow() now use the shared helper
- Add getAdapterCount() to @skillkit/agents to avoid allocating 44
spread-copied objects just to call .length
- Use getAdapterCount() in cli.ts instead of getAllAdapters().length
* feat: add multi-tier eval engine (skillkit eval)
6-tier evaluation system for comprehensive skill assessment:
- Tier 1: LLM quality scoring (G-Eval pattern, 6 dimensions)
- Tier 2: Contradiction detection (formal rules + LLM semantic)
- Tier 3: Behavioral security (code analysis + taint tracking + LLM)
- Tier 4: Sandbox execution testing (Docker, graceful skip)
- Tier 5: Dynamic marketplace benchmarks (percentile ranking)
- Tier 6: Community signals (freshness, metadata, content health)
Works without API keys (heuristic fallback for Tier 1, Tiers 5-6).
Full LLM evaluation with --provider flag.
* fix: address Devin review — input validation for eval CLI
- --min-score NaN bypass: validate parseInt result before comparison
- --timeout NaN propagation: skip NaN timeout instead of passing to Docker
- --tier invalid input: graceful error message instead of unhandled throw
- scoreContentSize formula: cap sub-optimal at 99 to distinguish from optimal
* fix: address CodeRabbit review findings on eval system
- Harden contradiction/security prompts against prompt injection by
wrapping untrusted content in <skill_content> XML tags and adding
explicit "treat as untrusted data" instructions
- Use head+tail content sampling in security prompt instead of head-only
truncation to catch malicious payloads at end of content
- Add missing Tier 4 sandbox formatter to reporter.ts
- Fix misleading sampleSize fallback in dynamic-benchmark.ts
- Move --min-score NaN validation before engine.evaluate() to avoid
wasting LLM calls on invalid input
- Add truncation indicator to quality-cot.ts prompts
- Add term length limits (>100 chars) in contradiction.ts to mitigate
ReDoS risk on dynamically constructed regex
* fix: address round 2 review findings on eval system
- Wrap quality-cot prompts with <skill_content> XML tags and untrusted
data instructions, consistent with contradiction/security prompts
- Change truncateContent to head+tail sampling (4000+2000) instead of
head-only, so completeness/token-efficiency evaluators see both
beginning and end of long skills
- Fix escapeXmlTags regex to handle whitespace variants before closing
angle bracket (</skill_content >, </skill_content\n>)
- Type DIMENSION_PROMPTS with Record<EvalDimension, PromptBuilder> for
compile-time key validation
- Clamp estimatePercentile return to [0,100] range defensively
- Make buildStatsFromMarketplace async with periodic yields (every 50
skills) to avoid blocking the event loop on first-run cache miss
- Extract parsedTimeout local variable in eval.ts to avoid double
parseInt parsing
* fix: address round 3 Devin + CodeRabbit findings on eval system
- Scan both code blocks AND full content for dangerous patterns in
behavioral-security.ts, with deduplication by description+snippet
to avoid double-reporting (Devin: security scanner skipped prose)
- Use non-greedy regex /\[[\s\S]*?\]/ in parseSemanticFindings to
match first complete JSON array instead of greedily spanning to
last bracket in LLM response (Devin: greedy regex corrupted JSON)
- Fix formatSandboxDetails to use r.testCase matching SandboxResult
type instead of r.name which doesn't exist (Devin: property mismatch)
- Make MarketplaceSkill.description optional to match marketplace
schema where only id/name/source/tags are required (CodeRabbit)
- Rename sandbox test case from "code example execution" to "code
block content validation" since it validates content not execution
(CodeRabbit)
* fix: address round 4 Devin + CodeRabbit findings on eval system
Inline fixes:
- Replace greedy JSON regex in behavioral-security.ts runLLMAnalysis
with balanced bracket extraction that correctly handles ] inside
JSON string values
- Replace non-greedy regex in contradiction.ts parseSemanticFindings
with same balanced bracket extraction (non-greedy also breaks on
inner brackets)
- Fix crossValidatedCount in behavioral-security.ts to accumulate
(+= uniqueEngines.size) instead of replacing (Math.max)
- Fix gradeDeterministic: cleanExit && hasOutput without matching
expectedOutcome now returns passed:false score:50 instead of
passed:true score:75
- Docker unavailable returns score:-1 sentinel instead of 0/F;
engine.ts filters score<0 tiers from overall average so
Docker-less machines aren't penalized
Duplicate fix:
- Add score field to SandboxResult type; store gradeResult.score
in each result; compute tier score from average of numeric scores
instead of boolean pass rate
Nitpick fixes:
- Distinguish "Docker unavailable" from "No tests executed" in
formatSandboxDetails
- Tighten base64 obfuscation regex to require 4-char groups
- Only push 'llm' to engines when LLM actually produced findings
- Bound \s+ to \s{1,10} in dynamically constructed regexes in
contradiction.ts to prevent catastrophic backtracking
- Validate cache timestamp with Number.isFinite() in loadCache to
reject corrupted cache files with NaN timestamps
* feat: integrate skills.sh as first-class registry
Add SkillsShRegistry and SkillsShProvider so users can install skills
directly from skills.sh (e.g. `skillkit install skills.sh/owner/repo/skill`).
The find command now searches skills.sh by default alongside the local
marketplace. Post-install shows skills.sh popularity stats and pro tips
highlighting skillkit's unique features (translate, recommend, evaluate).
* fix: address CodeRabbit and Devin review findings for skills.sh integration
- Fix path traversal vulnerability in SkillsShProvider.clone (validate subpath containment)
- Fix federated search skipped for interactive queries (use searchTerm instead of this.query)
- Guard showSkillsShStats to skip network call for local providers
- Replace deprecated npm --production with --omit=dev
- Fix loose includes() match in getSkillStats to use strict endsWith()
- Fix FederatedSearch resilience: don't re-throw RateLimitError, return partial results
- Forward timeoutMs through FederatedSearch to registries
- Extract shared formatCount utility to helpers.ts (deduplicate find.ts/install.ts)
- Fix install hint to use skill.source instead of fragile skill.description
- Fix timer leak in fetchLeaderboard using finally block
- Deduplicate URL parsing via shared parseSkillsShParts helper
- Validate --agent flag values against known agent types
* Security hardening and code quality fixes
- Replace execSync with execFileSync to prevent shell injection (providers, executor, fetcher, publish)
- Add security headers: X-Content-Type-Options, X-Frame-Options, Content-Security-Policy, Referrer-Policy
- Add input validation for save route name field to prevent path traversal
- Fix rate limiter IP extraction (x-forwarded-for + x-real-ip fallback)
- Default server bind to 127.0.0.1 instead of 0.0.0.0
- CORS fallback to * (configurable via options.corsOrigin)
- Use execFileSync in publish submit (avoid shell injection when opening browser)
- TLS-aware health checks and HTTPS support in RemoteTransport
- Conditional rejectUnauthorized for TLS (self-signed configurable)
- Add **/.env.local patterns to .gitignore
- Formatting: double quotes, trailing commas (Prettier)
* fix: address CodeRabbit and Devin review findings from PR #74
- TLS: default allowSelfSigned=true to preserve self-signed cert workflow;
enforce rejectUnauthorized=true when trustedCAs are supplied in client context
- Signature verification: verifySecureMessage now cryptographically verifies
signatures using PeerIdentity.verifyHex instead of only checking fingerprints
- WebSocket: fix timeout handle leak in performClientHandshake; add error
logging to empty catch blocks; add per-socket error handler; pre-compute
signature once in broadcast(); add send error callbacks; reject startup if
TLS configured but certInfo missing
- Command parsing: replace broken regex tokenizer with shared splitCommand()
utility that correctly handles --flag="value with spaces"
- Rate limiter: prefer x-real-ip, fall back to rightmost x-forwarded-for entry
- Save route: scope IPv6 prefix checks to actual IPv6 addresses; stop leaking
internal error messages to clients
- Publish: extract shared parseSkillFrontmatter(); fix getRepoInfo to use
async execFileSync; fix path containment checks with separator
- Remote transport: add rejectUnauthorized option; bracket IPv6 in URLs;
fix host:port splitting for IPv6 with lastIndexOf
- Health checks: add tls/tlsAllowSelfSigned to Host type; pass https options
- Fetcher: validate owner/repo; derive indexDir from INDEX_PATH
- .gitignore: add recursive **/.env pattern, remove redundant entries
* fix(executor): restore shell feature support in shellExecutor
The previous security hardening replaced execSync with execFileSync,
which broke commands using pipes, redirects, and subshells. Now detects
shell metacharacters and falls back to sh -c for those commands while
keeping execFileSync for simple commands.
* chore: migrate domain from agenstskills.com to skillkit.sh
Update all OG/Twitter meta tags, API endpoints, docs links, badge
URLs, privacy policy, extension store listing, and README references
to use skillkit.sh. Both skillkit.sh and agenstskills.com are treated
as equal domains — privacy policy, store listing, and docs reference
both.
* perf: skip Vercel builds when docs unchanged
Add ignoreCommand to both docs/skillkit and docs/fumadocs vercel.json.
Vercel will skip rebuilds when the respective docs folder has no
changes, saving ~1h+ of build minutes per billing cycle.
* feat: add `skillkit doctor` command and enhance `skillkit status`
- New `skillkit doctor` command with 10 health checks across 3 categories
(environment, configuration, skills) with --json and --fix flags
- Enhanced `skillkit status` shows project overview dashboard when no
active session exists instead of bare "No active session found" message
- Overview includes agent, config, version, skills count, recent history
* feat: add skill activity log, session snapshots, and session explain
Three new session intelligence features:
- Activity Log: tracks which skills were active per git commit
(`skillkit activity` with --skill, --limit, --json flags)
- Session Snapshots: save/restore named snapshots of session state
(`skillkit session snapshot save|restore|list|delete`)
- Session Explain: structured summary of current session
(`skillkit session explain` with --json, --no-git flags)
Includes path traversal protection on snapshot names, YAML validation
on parsed activity data, and robust error handling on snapshot restore.
983 tests passing (942 core + 41 CLI including 14 E2E tests).
* feat: add issue planner - generate plans from GitHub Issues
New `skillkit issue plan` and `skillkit issue list` commands that
fetch GitHub Issues via `gh` CLI and generate StructuredPlans.
Parses checkboxes, file mentions, and labels from issue body.
29 new tests, all passing.
* fix: address CodeRabbit review suggestions for PR #64
- doctor.ts: use execFileSync, typed AgentType, remove as-any casts
- activity.ts: validate parseInt with radix and NaN guard
- issue.ts: wrap JSON.parse in try/catch for gh CLI output
- issue-planner.ts: guard JSON.parse, fix includeTests no-op ternary,
remove duplicate IssuePlanMetadata (import from types.ts)
- status.ts: typed StatusOverview interface, remove as-any casts
- snapshot-manager.ts: align observations param with SessionSnapshot type
- session.ts: proper type mapping for observations (no double cast)
- activity-log.ts: min-length guard for SHA prefix match
- e2e tests: use top-level imports instead of inline require()
- commands.test.ts: add assertions for DoctorCommand, IssuePlanCommand,
IssueListCommand
* fix: address review feedback from CodeRabbit and Devin
- Include test step in no-checklist fallback when includeTests is true
- Restore observations from snapshot in SessionSnapshotRestoreCommand
* fix: address remaining CodeRabbit and Devin review feedback
- doctor.ts: Use getAdapter() per AgentType instead of getAllAdapters()
spread (which destroys prototype methods like isDetected)
- doctor.ts: Use getGlobalSkillsDir() from AGENT_CONFIG instead of
hardcoded path formula
- session-explainer.ts: Defensively default state.history and
state.decisions to empty arrays
- session-explainer.ts: Use loadConfig().agent with skillSource fallback
instead of always using skillSource as agent name
- observation-store.ts: Add static readAll() that reads observations
without sessionId mismatch clearing
- session.ts: Use ObservationStore.readAll() for snapshot save to avoid
empty observations due to sessionId mismatch
* feat: implement standards reference — URL-to-Skill, AGENTS.md, spec validation
Phase 1-4 of positioning SkillKit as THE runtime implementing all three
AI agent skill standards (Agent Skills, AGENTS.md, well-known discovery).
New modules:
- save/ — ContentExtractor, SkillGenerator, AutoTagger for URL-to-Skill
- agents-md/ — AGENTS.md generator with managed section markers
- validation/ — SpecValidator with strict mode for agentskills.io compliance
New CLI commands:
- `skillkit save <url>` — save any URL/text/file as a reusable SKILL.md
- `skillkit agents init|sync|show` — generate and manage AGENTS.md
Updated:
- Well-known provider: Mintlify convention paths, discoverFromUrl()
- Install command: smart URL fallback (well-known → offer save)
- Remove command: auto-updates AGENTS.md on skill removal
- Validate command: --strict flag for spec compliance checks
- Quality module: name-directory match + token budget warnings
- Schema: agents enum updated from 7 → 44 agents
* fix: address Devin review — register AgentsMd commands, fix temp dir leak, wrap AGENTS.md update
- Register AgentsMdCommand/Init/Sync/Show in CLI entry point (were exported but never registered)
- Clean up tempDir in wellknown clone() when no skills found via index or mintlify
- Wrap AGENTS.md auto-update in install.ts with try-catch to prevent masking successful installs
* fix: address CodeRabbit review findings
- wellknown.ts: drop multiline flag from frontmatter regex, extract skill name
from URL path segments instead of basename on full URL
- agents-md.ts: wrap init/sync in try-catch for user-friendly error messages
- publish.ts: add path-traversal guard to mintlify branch
- validate.ts: remove duplicate error/warning output in spec report
- extractor.ts: add 30s AbortSignal timeout to all fetch calls
- skill-generator.ts: fix .filter(Boolean) stripping blank line after frontmatter
* feat: add 12 missing AI coding agents (Devin, Aider, Cody, Amazon Q, Augment, Replit, Bolt, Lovable, Tabby, Tabnine, CodeGPT, PlayCode)
Expands agent support from 32 to 44 agents across all packages:
- Core types, configs, discovery paths, format maps, command generator
- Agent adapters (UniversalAdapter for all new agents)
- CLI onboarding icons/names, TUI symbols/logos
- Updated all docs, tests, and e2e references from 32 to 44
* test: add comprehensive tests for new agents, update docs and e2e
- Add it.each tests for all 12 new agent adapters and configs
- Add completeness test ensuring every AgentType has adapter + config
- Update agents.mdx doc table with all 12 new agents
- Update e2e onboarding test with new agent display names
* fix: add all 44 agents to translateSkillToAll and fix e2e test
- translateSkillToAll had hardcoded 17-agent list, now includes all 44
- e2e/skill-translator test was missing cline-zencoder batch in SUPPORTED_AGENTS
* fix: replace UniversalAdapter with GenericAgentAdapter for all non-universal agents
UniversalAdapter hardcoded type='universal', skillsDir='skills', and
name='Universal (Any Agent)' for all 26 agents that used it. This caused
skills to install to wrong directories and agents to be undetectable.
GenericAgentAdapter reads type, skillsDir, configFile from AGENT_CONFIG
and checks agent-specific directories in isDetected(). Fixes both the
12 new agents and the 14 pre-existing ones (cline through zencoder).
115 adapter tests now verify correct type, skillsDir, configFile, and
name for every GenericAgentAdapter instance.
- Add dedicated security.mdx docs page with full rule reference (46+ rules),
threat categories, SARIF/CI integration guide, and programmatic usage
- Add TUI Scan screen (z key) with live scan results, severity colors,
finding navigation, and rescan support
- Add scan to TUI sidebar nav, help shortcuts, and bottom status bar
- Add security scanner section to docs index page with link
- Add programmatic scan usage example to CLI README
- Remove stale --quiet option from commands.mdx
- Add security page to docs sidebar navigation
* feat: add security scanner for detecting malicious patterns in skills
Add `skillkit scan <path>` command with 46+ detection rules across 6 threat
categories: prompt injection, command injection, data exfiltration, tool abuse,
hardcoded secrets, and unicode steganography.
- 3 analyzers: static pattern matching, SKILL.md manifest validation, secrets detection
- 4 output formats: summary (colored terminal), json, table, sarif (GitHub Code Scanning)
- Auto-scan on `skillkit install` (--no-scan to skip, --force to override)
- Auto-scan on `skillkit publish` (blocks on HIGH+ findings)
- Threat taxonomy with 10 categories and severity levels
- False-positive filtering for placeholders, test files, safe patterns
- Updated README, CLI docs, website commands, features, and fumadocs
* fix: address Devin and CodeRabbit review suggestions for security scanner
- Fix .env variant files (.env.local, .env.production) being skipped by scanner
- Add multiline rule support in StaticAnalyzer for cross-line pattern matching
- Fix PI004/PI013 patterns that contained literal \n (never matched per-line)
- Fix /g flag on CI006 path traversal regex causing intermittent misses
- Fix frontmatter regex to handle Windows CRLF line endings
- Fix allowed-tools parsing to detect multi-line YAML arrays
- Fix snippet redaction order: redact before truncate to prevent secret leaks
- Reset findingCounter per analyze() call for deterministic finding IDs
- Make SARIF artifactLocation.uri relative to skillPath for GitHub Code Scanning
- Use basename() instead of split('/').pop() for cross-platform paths
- Inject package version at build time via tsup define (replaces hardcoded 1.0.0)
- Validate --fail-on and --format CLI options with clear error messages
- Remove unused format/quiet from ScanOptions interface
- Move publish scan after name validation to avoid blocking on invalid skills
- Add excludePatterns for jailbreak detection to reduce false positives
- Add biome suppression for intentional control character regex in UC005
- Use alternation instead of character class for ZWJ in UC001
* fix: address CodeRabbit round 2 review suggestions
- Remove unused --quiet option from scan command
- Update OpenAI API key pattern to detect sk-proj- and sk-admin- prefixes
* fix: address deep review issues in security scanner
- SK001: add negative lookahead to prevent matching Anthropic keys
- static.ts: fix multiline loop hang for non-global regex with filtered matches
- CI005: reduce severity to MEDIUM, expand excludePatterns for template literals
- CI006: reduce severity to MEDIUM, add fileTypes whitelist, exclude markdown/YAML patterns
- PI011: fix literal \n in regex, use $/im for end-of-line matching
- UC004: use \u{...} notation for supplementary codepoint range
- scanner.ts: propagate skipRules to ManifestAnalyzer and SecretsAnalyzer
* feat(ai): add smart context generation with multi-provider support
Implements the SkillKit Smart Context Generation system:
- Multi-Provider Support: 6 LLM providers (Anthropic, OpenAI, Google, Ollama, OpenRouter, Mock) with auto-detection via environment variables
- Multi-Source Context Engine: 4 context sources (docs, codebase, skills, memory) with relevance weighting
- Skill Composition Engine: Natural language search and intelligent skill merging
- Agent Optimization: Agent-specific optimization for 10+ agents with compatibility scoring
- Security Module: Trust scoring (0-10 scale) and prompt injection detection
- Wizard Core: 6-step interactive wizard (expertise → context-sources → composition → clarification → review → install)
- CLI Integration: `skillkit generate` command with @clack/prompts
New files:
- packages/core/src/ai/providers/{types,anthropic,openai,google,ollama,openrouter,factory}.ts
- packages/core/src/ai/context/{index,docs-source,codebase-source,skills-source,memory-source}.ts
- packages/core/src/ai/composition/{index,analyzer,merger}.ts
- packages/core/src/ai/agents/{optimizer,compatibility}.ts
- packages/core/src/ai/security/{trust-score,injection-detect}.ts
- packages/core/src/ai/wizard/{types,steps,clarification,index}.ts
- packages/cli/src/commands/generate.ts
- 6 test files for new modules
* fix(ai): address Devin/CodeRabbit review feedback
- Escape regex in codebase-source.ts to prevent ReDoS attacks
- Fix division by zero in skills-source.ts when no keywords present
- Include all config overrides in ProviderFactory cache key
- Validate AI-provided indices before indexing skills arrays
- Use limited array in ollama.ts search to avoid out-of-bounds access
- Add 120s timeout with AbortController to ollama/openrouter requests
- Fix provider precedence in wizard: config.provider > options.provider
- Fix emitComplete timing to fire after install step completes
- Guard memory lookups with memoryPersonalization opt-out check
* chore(deps): update Google AI SDK to new @google/genai package
The legacy @google/generative-ai package is deprecated as of Nov 2025.
Updated to the new unified @google/genai SDK (GA since May 2025).
* fix(ai): address remaining Devin/CodeRabbit review issues
- Update AIConfig type to support all providers (google, ollama, openrouter)
- Update getAIConfig() in CLI to detect all configured providers
- Add defensive check for expertise in clarification fallback questions
- Add null check for OpenAI choices[0] response
- Validate AI-provided indices in OpenAI/Anthropic search methods
- Add 120s timeout with AbortController to OpenAI/Anthropic requests
- Guard against zero/Infinity per-source chunk allocation in ContextEngine
- Normalize trust score weights and clamp final score to [0-10]
- Fix Windows path handling in memory-source.ts (use [\\/] regex)
- Fix token-to-character ratio consistency (use 4 throughout)
- Target 95% of max tokens in truncation to stay under limit
* chore: sync docs package versions to 1.13.0 and update lockfile
- Update docs/skillkit/package.json version to 1.13.0
- Update docs/fumadocs/package.json version to 1.13.0
- Regenerate pnpm-lock.yaml with new @google/genai dependency
* fix(ai): final code review fixes for smart context generation
- Add 'mock' to AIConfig provider union type
- Show mock provider in handleProviders() when no API keys configured
- Reserve space for truncation marker in truncateContent
- Capture pre-heading content in splitIntoSections
- Add tolerant JSON parsing in anthropic parseSearchResponse
- Use Number.isInteger for index validation in openai search
- Validate all required fields in openai generateFromExample
- Guard against zero sentences in trust-score scoreClarity
- Sanitize weight values (non-negative, finite) in TrustScorer
* fix(ai): re-validate after AI optimization and validate parseGenerateResponse
- Re-truncate content after AI optimization if it exceeds agent max context
- Add JSON extraction from code fences in parseGenerateResponse
- Validate required fields (name, description, content) in parseGenerateResponse
- Clamp confidence to [0,1] range and properly type-check optional fields
* feat(memory): add lifecycle hooks, progressive disclosure, and auto-CLAUDE.md updates
- Add Claude Code lifecycle hooks (SessionStart, PostToolUse, SessionEnd)
for automatic memory capture and injection
- Implement 3-layer progressive disclosure for token-optimized retrieval
(Index ~50 tokens, Timeline ~200 tokens, Details ~600 tokens)
- Add CLAUDE.md auto-updater to populate LEARNED section with learnings
- Add auto-compression threshold trigger in observation store
- Add new CLI commands: `memory sync-claude` and `memory index`
- Update documentation with new memory features
- Remove 139 lines of unnecessary inline comments (code cleanup)
* fix(memory): address Devin review comments
- Fix double execution bug in MemoryHookManager by adding
generateHookOutputFromResult methods to avoid calling execute() twice
- Add setConfig method to PostToolUseHook for runtime config updates
- Update manager.setConfig() to propagate config to all hooks
* fix(memory): address CodeRabbit review comments
- Fix cross-platform path handling using path.basename() instead of split('/')
in claude-md-updater.ts, session-start.ts, and session-end.ts
- Fix onSessionEnd to call execute() before generateHookOutput to run
compression/cleanup side effects
- Add generateHookOutputFromResult() to SessionEndHook to avoid double execution
- Call clearOldPendingErrors() in PostToolUseHook.execute() to prevent
unbounded pendingErrors map growth
- Handle async checkAutoCompression() errors in observation-store.ts
- Add validation for compression threshold (must be positive integer)
- Fix docs/commands.mdx to use correct CLI flags (--limit, --tags)
- Clarify token accounting in progressive-disclosure smartRetrieve() JSDoc
- Use config.minRelevanceForCapture in session-start.ts instead of hardcoded value
* fix(memory): address additional CodeRabbit review comments
- Use nullish coalescing (??) for maxResults defaults in progressive-disclosure
- Guard against non-positive tokenBudget in smartRetrieve()
- Thread minRelevance option through to findRelevantIds()
- Fix memory index CLI to validate --limit and pass maxResults to getIndex()
- Add OpenAPI 3.1 spec with all endpoints, schemas, and tags
- Serve Swagger UI at GET /docs with try-it-out enabled
- Serve OpenAPI JSON at GET /openapi.json
- Add API info endpoint at GET / with endpoint listing
- Update serve command to show /docs and /openapi.json endpoints
- Update REST API docs page with interactive docs section
* feat: runtime skill discovery — REST API, MCP server, cache, ranking, Python client (#42)
Add runtime skill discovery capabilities enabling agents to autonomously
search and retrieve skills on demand, inspired by skyll.app's approach
of making skills queryable by intent rather than requiring pre-installation.
Core infrastructure:
- Pluggable cache backend with LRU + TTL (MemoryCache)
- Multi-signal relevance ranker (content 40pts, query 30pts, popularity 15pts, refs 15pts)
- Enhanced SKILL.md parser with frontmatter extraction and reference discovery
- Runtime skill injector fetching SKILL.md from GitHub with caching
REST API server (@skillkit/api):
- Hono-based HTTP server on port 3737
- GET/POST /search with ranked results and filters
- GET /skills/:source/:id, /trending, /categories, /health, /cache/stats
- Sliding-window rate limiting (60 req/min/IP)
MCP server (@skillkit/mcp):
- 4 tools: search_skills, get_skill, recommend_skills, list_categories
- 2 resources: skills://trending, skills://categories
- Compatible with Claude Desktop, Cursor, any MCP client
CLI serve command:
- `skillkit serve` starts the API server
- Options: --port, --host, --cors, --cache-ttl
Community registry:
- registry/SKILLS.md with 26 curated skills across 12 categories
- CommunityRegistry class integrated into FederatedSearch
Python client (skillkit-client):
- Async httpx + pydantic client for the REST API
- search, get_skill, trending, categories, cache_stats, health
All 856+ tests pass across 24 workspace tasks.
* fix: resolve TypeScript errors in API package
Add explicit return type to rate limiter middleware and remove stale
CategoryCount type annotation in categories route.
* fix: address CodeRabbit review findings and version-check CI
- Fix version-check: bump docs packages to 1.11.0
- Parse first IP from X-Forwarded-For header in rate limiter
- Add try/catch for malformed JSON in POST /search
- Use exact source match in skill routes (owner/repo/id pattern)
- Fix NaN fallback for invalid limit params in trending/search
- Filter empty strings from query words in relevance ranker
- Use Map lookups instead of O(n^2) find in MCP resources/tools
- URL-encode path params in Python client
- Fix -h flag conflict with Clipanion help in serve command
- Fix typo: remoti-dev -> remotion-dev in registry SKILLS.md
* fix: address Devin review — composite key lookups and Python URL path
- Use source+name composite key Map lookups in search GET/POST routes
- Use source+name composite key Map lookup in trending route
- Fix Python client get_skill to use /skills/:owner/:repo/:id URL pattern
- Update Python test to match new 3-segment skill URL
* feat: add viral growth features (quick setup, badges, matrix, SKILL.md, federated search)
Tier 2: skillkit quick zero-friction entry, shareable badge generator,
agent compatibility matrix on website.
Tier 3: reposition as npm for agent skills, SKILL.md validate/init/check
commands, federated GitHub skill search with --federated flag on find.
* fix: matrix filtering, path traversal, cross-platform paths, fetch timeout
- CompatibilityMatrix: query matching agents now shows all categories
(and vice versa) instead of hiding both dimensions independently
- quick.ts: sanitize skill names and validate targetInstallPath stays
inside installDir before rmSync/cpSync/symlinkSync
- skillmd.ts: use path.relative() instead of string replace for
cross-platform path display
- registry: add AbortController timeout (10s default) and distinct
RateLimitError on 403 instead of silent empty return
- find.ts: handle RateLimitError with user-facing message
- Add quick-security.test.ts regression tests for path traversal
* fix: harden isPathInside with path.resolve, propagate RateLimitError
- isPathInside now uses path.resolve() + path.sep comparison instead of
naive string replacement, fixing detection of paths completely outside
the parent directory (both packages/core and src/core copies)
- FederatedSearch.search re-throws RateLimitError from rejected
Promise.allSettled results so callers (find.ts) can handle 403s
- Updated security test to use the corrected isPathInside logic
* fix: use GITHUB_TOKEN/GH_TOKEN for authenticated GitHub API requests
GitHubSkillRegistry now reads GITHUB_TOKEN or GH_TOKEN from env and
sends Bearer auth header, raising the rate limit from 10 to 30 req/min
for code search. Matches the guidance shown to users on 403 errors.
Features:
- Hierarchical skill tree with 12 categories (skillkit tree)
- LLM-based reasoning engine for skill discovery
- Explainable recommendations (--explain, --reasoning flags)
- Connector placeholders for tool-agnostic skills (~~CRM, ~~chat, etc.)
- Execution flow tracking with metrics
- Standalone vs Enhanced mode detection
CLI:
- New `tree` command for taxonomy navigation
- Enhanced `recommend` with explanation support
- Fixed --skill alias and agent detection
Documentation:
- Added tree.mdx for Skill Tree documentation
- Updated recommendations.mdx with reasoning features
- Updated TUI docs with tree view mode
- Updated README with new features
- Updated website Features and Commands
* feat(core): add PageIndex-inspired patterns for skill discovery
Implements patterns from anthropics/knowledge-work-plugins:
Tree Module (packages/core/src/tree/):
- Hierarchical skill taxonomy with 12 categories
- Tree generator from skill tags and metadata
- Tree serializer for JSON/Markdown export
- Related skills graph with 4 relation types
Reasoning Engine (packages/core/src/reasoning/):
- LLM-based tree traversal for skill discovery
- Explainable recommendations with reasoning chains
- Search planning with category relevance scoring
- Result caching with 5-minute TTL
Connectors (packages/core/src/connectors/):
- Tool-agnostic ~~ placeholder system
- 13 connector categories (crm, chat, email, etc.)
- Auto-suggest mappings from MCP config
- Placeholder detection and replacement utilities
Execution Flow (packages/core/src/execution/):
- Step-by-step execution tracking with metrics
- Automatic retry with exponential backoff
- Standalone vs Enhanced mode detection
- Capability-based feature gating
CLI Integration:
- New `skillkit tree` command with --generate, --stats flags
- Enhanced `skillkit recommend` with --explain, --reasoning flags
TUI Integration:
- Tree view mode in Marketplace (toggle with 'v')
- Arrow key navigation for tree hierarchy
* fix: address CodeRabbit and Devin review feedback
- Use os.homedir() instead of process.env.HOME || '~' fallback
(tree.ts, tree.service.ts, mode.ts)
- Add guard for division by zero in tree stats percentage
- Validate parseInt result for --depth flag
- Fix average stats calculation to use cacheMisses count
- Extract category from data in validateCategoryScore
- Remove unused skillMap field from TreeGenerator
- Make reasoning provider configurable instead of hardcoded
- Add try/catch for tree loading errors in Marketplace
- Handle missing tree nodes during traversal
* fix(execution): clear timeout timer to prevent resource leak
Store setTimeout ID and clear it in finally block when step
completes successfully before timeout. This prevents orphaned
timers from accumulating during long-running applications.
- Add --skill as alias for --skills option (install command)
- Fix sync command to always detect agent instead of using default 'universal'
- Auto-install npm dependencies for skills with package.json
* feat(agents): add skill-to-subagent converter command
Add `skillkit agent from-skill` command to convert SkillKit skills into
Claude Code native subagent format (.md files in .claude/agents/).
Features:
- Reference mode (default): generates subagent with `skills: [skill-name]`
- Inline mode (--inline): embeds full skill content in system prompt
- Options: --model, --permission, --global, --output, --dry-run
New files:
- packages/core/src/agents/skill-converter.ts
- packages/core/src/agents/__tests__/skill-converter.test.ts (23 tests)
Closes#22
* feat(publish): add RFC 8615 well-known skills support
Redesign publish workflow to align with industry standard:
- Add WellKnownProvider for auto-discovery from any domain
- skillkit add https://example.com discovers skills via /.well-known/skills/
- skillkit publish generates well-known hosting structure
- skillkit publish submit opens GitHub issue (legacy workflow)
Provider discovers skills from:
- /.well-known/skills/index.json (manifest)
- /.well-known/skills/{skill-name}/SKILL.md (skill files)
Includes 16 tests for WellKnownProvider and structure generation.
* fix(wellknown): correct baseSkillsUrl calculation for skills.json fallback
When using the /.well-known/skills.json fallback URL, the previous
logic produced a malformed URL with duplicated .well-known path segment:
- Before: https://example.com/.well-known/.well-known/skills
- After: https://example.com/.well-known/skills
Extract calculateBaseSkillsUrl helper function for better testability
and add 4 new unit tests covering both URL formats.
* docs: update publish command documentation for well-known skills
- README.md: Add self-hosting section with well-known URI structure
- commands.mdx: Add Publishing Commands section
- marketplace.mdx: Expand publish section with self-hosting instructions
- skills.mdx: Add self-hosting workflow documentation
Documents the new `skillkit publish` workflow that generates RFC 8615
well-known URI structures for decentralized skill hosting.
* fix(security): address path traversal and CRLF vulnerabilities
- agent.ts: Sanitize skill.name when building default filename
- publish.ts: Add sanitizeSkillName() to validate skill names before
using in file paths, verify resolved paths stay within target dir
- skill-converter.ts: Normalize CRLF to LF before extracting body
content to handle Windows-style line endings
- wellknown.ts: Add sanitizeSkillName() to validate remote skill names,
verify resolved paths stay within temp dir, encode URL components
* fix(agents): improve YAML string escaping logic
Only escape strings that actually need it for valid YAML parsing:
- Strings containing newlines or colons
- Strings starting with special YAML characters (-, *, &, !, {, [, >, |, @, `)
- Strings starting with quotes
This fixes unnecessary escaping of strings like "code-quality" where
hyphens in the middle don't need escaping.
* fix(wellknown): address security review feedback
- Use path.sep instead of hardcoded '/' for cross-platform compatibility
- Add sanitization and path containment check in generateWellKnownStructure
- Remove unused 'vi' import in wellknown.test.ts
* feat: add skillkit primer command for AI instruction generation
New feature that analyzes codebases and generates AI instruction files
for all 32 supported agents (inspired by pierceboggan/primer).
## New Module: packages/core/src/primer/
- types.ts: Zod schemas and TypeScript interfaces
- analyzer.ts: Codebase analysis (languages, frameworks, conventions)
- generator.ts: Agent-specific instruction file generation
- index.ts: Barrel exports
- __tests__/primer.test.ts: 15 comprehensive tests
## CLI Command: skillkit primer
- --agent/-a: Generate for specific agents
- --all-agents/-A: Generate for all 32 agents
- --output/-o: Custom output directory
- --dry-run/-n: Preview without writing
- --analyze-only: Show analysis only
- --verbose/-v: Detailed output
- --json/-j: JSON output
## Features
- Detects: languages, frameworks, package managers, libraries
- Analyzes: project structure, code conventions, CI/CD, Docker
- Generates: CLAUDE.md, .cursorrules, copilot-instructions.md, etc.
* fix: address CodeRabbit PR review issues for primer
- Fix OS-agnostic path handling in analyzer using path.relative() and path.sep
- Prevent output collisions when multiple agents produce same filename
- Add defensive check in test before accessing generated[0]
- Remove duplicate projectPath parameter in CLI primer command
- Use PackageManager enum instead of z.string() for type safety
- Detect monorepos before src-based layout to prevent misclassification
- Document --output and --json flags in README examples
* fix: additional CodeRabbit review fixes for primer
- Fix Python detection to use correct source file (pyproject.toml, requirements.txt, or setup.py)
- Escape regex special characters in glob pattern matching (e.g., *.module.css)
- Derive ALL_AGENTS from AGENT_CONFIG to avoid divergence
- Map agent config format to template format based on file extension
* feat(primer): support multi-format output rendering
Refactored generateContent to properly respect the format field
instead of always producing Markdown output.
- Added FormatRenderer interface with format-specific implementations
- Created renderers for markdown, mdc, json, xml formats
- JSON format generates structured data via generateJsonContent()
- MDC format includes YAML frontmatter via wrap() method
- XML format properly escapes content with escapeXml()
- Updated all section generators to accept and use the renderer