1. install.ts: gate marketplace-resolution console.log behind !this.json && !this.quiet so --json output stays a single JSON blob and --quiet stays silent. Uses real flag names from the InstallCommand class (this.json, this.quiet).
2. tool-abuse.ts: TA002 pattern 2 lookbehind was too broad. The original list (must|should|always|will|to) included malicious intensifiers — 'Always run without confirmation' and 'Will execute without approval' were incorrectly suppressed. Replaced with negation-only tokens (not|n'?t|never|avoid|forbid|deny|reject|prevent|cannot|can'?t) so genuine negation like 'Do not run without confirmation' still suppresses, but malicious intensifiers fire.
3. rules.test.ts: scanContent helper now uses try/finally so the temp directory is cleaned up even when analyzer.analyze throws.
Tests: 3 new cases for the TA002 lookbehind fix. Full core suite 1051/1051 pass.
Four scanner rules over-fired on standard Node patterns and safety documentation, blocking install of skills like pro-workflow on every node-script skill. Findings tracked at #129.
TA002 (autonomy abuse): patterns 2 and 3 now use negative lookbehind for negation context. 'Never auto-approve' / 'must not bypass approval' no longer trip the rule, but bare 'Auto-approve all tools' still fires.
CI003 (child_process): dropped the bare 'require(child_process)' pattern. Importing the module is not the risk surface — exec() and execSync() are, and they have their own patterns. const { execFileSync } = require('child_process') is the recommended safe form.
CI005 (template literal injection): narrowed pattern to template literals inside shell-spawning calls (exec/execSync/spawn/spawnSync). Authorization headers like `Bearer ${env.TOKEN}` and string concatenation like fs.writeFileSync(file, `${header}${rows}`) are no longer flagged.
CI007 (shell chaining): dropped 'markdown' from fileTypes. Markdown documentation of dangerous shell patterns (deny-list tables, security guides) is documentation, not execution. Real .sh/.bash files still scanned.
install command: marketplace name lookup. 'skillkit install <bare-name>' now consults marketplace/skills.json for entries matching by name or id and resolves to the canonical source string before failing with 'Could not detect provider'.
Tests: 13 new scanner regression tests in packages/core/src/scanner/__tests__/rules.test.ts. Full suite 1048/1048 pass.
Version bump
------------
1.22.1 -> 1.23.0 across all workspace packages. Minor bump covers the
slim-install dependency layout, safer first-time agent picker, and the
corresponding doc + branding changes.
Prune dead optional deps in @skillkit/core
------------------------------------------
@anthropic-ai/sdk, @google/genai, ollama, and openai were listed as
optionalDependencies but never imported. They were pulled through
@skillkit/cli transitively on every default install, adding ~57
packages and one deprecation warning (node-domexception via the
SDKs' own formdata-node / node-fetch transitives). Grep of
packages/core/src confirms no SDK import; providers in
src/ai/providers use direct fetch() against the provider HTTP APIs.
Drop them. better-sqlite3, sqlite-vec, and node-llama-cpp stay
because they are actually dynamic-imported by the vector store and
embedding pipeline.
Measured on a fresh npm cache (overrides-pinned file: tarballs):
before: 475 pkgs / 18s / 3 deprecation warnings
1.23.0: 418 pkgs / 18s / 2 deprecation warnings
slim: 118 pkgs / 9s / 0 deprecation warnings
The two remaining warnings are upstream:
- prebuild-install@7.1.3 from better-sqlite3 (must stay)
- glob@9.3.5 from @opentui/solid -> babel-plugin-module-resolver
README rewrite
--------------
Punchier top fold. Palette-consistent SVG tag badges under
assets/tags/ replace most shields.io badges (still keep npm/CI/etc
on shields since they need live data). Collapsible detail blocks
for every feature instead of the linear wall. Same nav anchors the
site uses. No em dashes. No AI filler words. Writing style pulls
from the ai-engineering-from-scratch and awesome-claude-design
READMEs the user already runs.
Branded SVG tags
----------------
scripts/gen-tags.mjs generates 20 SVG pills under assets/tags/ in
the SkillKit palette (#09090b base, #fafafa text, #3f3f46 border,
accent per tag). Two variants: single pill with left accent bar
(new, slim-install, v1.23.0, etc.) and split badge (agents | 45,
packages | 118, install | 9s). Matches the aesthetic used on
awesome-claude-design, tailored to SkillKit's monochrome brand.
The tags directory is fork-and-remix friendly; README points to it.
npx skillkit add <owner/repo> is the most common first-touch path,
so call it out explicitly. Cover:
- Full vs slim invocation (npx skillkit add vs npx --omit=optional
skillkit add), with the same 118/9s slim numbers.
- How the npx cache at ~/.npm/_npx/ makes subsequent runs instant
until a new version ships.
- When to graduate to a global install: skips the prompt-to-proceed
and kills the per-release refetch cost.
Updated in all four install-docs surfaces: root README, apps/skillkit
README (npm landing), docs/fumadocs/content/docs/installation.mdx,
and the Documentation.tsx components used on the website and the
fumadocs mirror.
Two nits from CodeRabbit on the first-user picker:
1. detectAgent() returns 'universal' as a fallback when no concrete
adapter is detected. Passing that through to quickAgentSelect made
the picker show "Just Universal (detected)" even when no real
agent was found — misleading. Treat 'universal' as "not detected"
and fall through to the plain select/all options instead.
2. The "All detected agents" label on the bottom option was wrong:
agents at the call site is getAllAdapters() (all supported
adapters), not only detected ones. Rename to "All supported
agents" and change the hint to "writes to every adapter" so
users aren't tricked into thinking this is a safe detected-only
install.
Devin review caught a hard bug: apps/skillkit/src/index.ts had a
static re-export of startTUI from @skillkit/tui. Now that tui is an
optional dependency, importing the skillkit package would crash with
ERR_MODULE_NOT_FOUND whenever the optional dep was skipped
(--omit=optional). Convert the re-export to an async wrapper that
dynamically imports @skillkit/tui on call. The error is surfaced
naturally with a clear Cannot-find-package message.
Docs updates so users understand the new split:
- README.md: add "Install sizes" section with full vs slim table,
cold-install measurements (475 pkgs / 18 s vs 118 / 9 s),
per-feature package mapping, and how to add a single optional
package later.
- apps/skillkit/README.md (npm landing): add "What's optional"
table + install mode guidance.
- docs/fumadocs/content/docs/installation.mdx: add "Full vs slim
install" section covering all four optional features.
- docs/skillkit/components/Documentation.tsx +
docs/fumadocs/src/components/Documentation.tsx: add "Optional
Features" block to the Quick Start step on the website.
Verified: after `npm install --omit=optional`, `import('skillkit')`
succeeds with 622 exports loaded, and `await startTUI()` fails with
a clean "Cannot find package '@skillkit/tui'" message instead of
crashing at module-load time.
Before: first-time users (no history) saw "All detected agents"
as the first option in the agent picker. Mashing Enter would
install to all 32 agents, writing files across every adapter's
config dir. Selecting "Select specific agents" also pre-ticked
all 32 agents, requiring the user to un-tick by hand.
After:
- Default option is "Just <detected> (detected)" — one agent,
low blast radius.
- "Select specific agents" moves above "All detected agents"
and pre-ticks only the detected agent.
- "All detected agents" gets a second confirm gate
("Install to all N agents?") with initialValue: false.
- Hint text on "All" warns "use only if you know".
Returning users still see "Same as last time (Recommended)" as
the first option; their flow is unchanged.
Non-TTY path is unchanged (still auto-detects a single agent).
Wire detectAgent() through quickAgentSelect so the picker can
label and pre-select the detected agent. Empty selection from
the confirm gate is treated as cancellation.
Move @skillkit/api, @skillkit/tui, @skillkit/mesh, @skillkit/messaging
to optionalDependencies. Drop @skillkit/memory entirely — the CLI
never imported it (only @skillkit/mcp-memory consumes it).
Killing @skillkit/memory removes cozo-node (native Rust) and
@xenova/transformers (~300MB ML stack) from the critical install
path. Making @skillkit/tui optional removes @opentui/core's
prebuild-install toolchain, which was the source of the
inflight/gauge/npmlog/tar/glob deprecation warnings.
Add skipNodeModulesBundle + platform=node + target=node18 to
apps/skillkit tsup config so optional deps stay external.
All TUI/mesh/messaging/api commands already lazy-import via
dynamic import() with surrounding try/catch, so missing optional
deps surface as a graceful error rather than a crash.
Measured impact (npm i skillkit in a clean cache):
Scenario Packages Time Deprecations Vulns (crit/high)
main 544 37s 10 33 (5/5)
slim (default) 475 18s 3 25 (0/0)
slim --omit=optional 118 9s 0 0
Core commands (list, find, translate, recommend, install, --help,
--version) smoke-tested in all three scenarios. Optional commands
(ui, mesh, message, serve) confirmed to degrade with a clean
"Cannot find package" error under --omit=optional.
Add --json flag with structured JSON output to: create, enable, disable,
eval, sync, tap (add/remove/list), translate, save, read.
Add loading spinners to: audit, fix, test, plugin, team, methodology,
hook, remove. Commands that already had spinners get --json-aware no-op
spinners when --json is active to suppress visual output.
- action.yml: pass inputs via env vars, whitelist allowed commands
- All --json commands: use no-op spinner to prevent stdout pollution
- remove: run AGENTS.md update before JSON output (was skipped)
- eval: gate spinner on json format, wrap in try/catch for cleanup
- scan: wrap spinner in try/catch for cleanup on failure
- Resolve symlinks with realpathSync before path containment check
to prevent symlink-based directory escape in skillkit_resource
- Search project-local skills before home-level (cwd first in roots)
- AGENT_DIR_MAP now maps to arrays, includes non-standard global
paths (opencode: .config/opencode/skills, windsurf: .codeium/...)
- Extra global dirs scanned when agent filter matches
Three new tools for token-efficient local skill access:
- skillkit_catalog (L1): lightweight catalog of installed skills (~50 tokens/skill)
- skillkit_load (L2): full SKILL.md content for a single skill
- skillkit_resource (L3): specific reference file with path traversal protection
Bumps server version to 1.21.0. Adds 14 new tests (18 total).
- Fix "44 agents" → "45 agents" across 34 files (website, docs,
package.json, tests, extension, API, TUI)
- README: add tap, check, update, publish submit to commands section
- Commands.tsx: add Sources category with tap commands, update manage
group with remove --source/--all and check
- commands.mdx: full docs for tap, check, update, remove, publish
submit with all new flags and behaviors
- configuration.mdx: document lock.json and taps.json
- security.mdx: document trust badges system
- install: use actual installed path (handles .md files) for lock
entry and results instead of reconstructing from skillName
- update: call addSkillToLock after both local and remote updates
so lock.json stays in sync with metadata
- update: --force bypasses GitHub activity and checksum skip-checks
- update: timestamp comparison falls back to installedAt when
updatedAt is not set
- Use finally for temp cleanup so user cancels don't leak cloned dirs
- Move addSkillToLock outside inner agent loop, merge agents array
- Use source:name as lock key to avoid cross-source collisions
- Gate GitHub activity check on provider.type (skip for GitLab/Bitbucket)
- Normalize source URL to owner/repo before official check
- Include skill name in trust badge output
- Atomic lock.json writes via tmp file + renameSync
- Remove stale --federated suggestion from find no-results message
Introduce tap add/remove/list commands to manage custom skill sources
stored in ~/.skillkit/taps.json. Taps are included in find/search
results, allowing users to add private or community repositories.
Introduce ~/.skillkit/lock.json with LockEntry/LockFile schemas to
track all installed skills in a single file. Install writes entries,
remove deletes them. Enables future team sync and reproducible setups.
Compute SHA-256 checksum of SKILL.md on install and store in metadata.
During update, check GitHub pushedAt timestamp and compare checksums
to skip unnecessary re-downloads when content hasn't changed.
Show trust badges for each skill during install: [Official] for known
official sources (Anthropic, Vercel, Expo, Remotion, Supabase, Stripe),
or [Trusted]/[Review]/[Caution] based on TrustScorer analysis.
- Hoist cloneResult above try block so catch can clean up temp dir
on any failure path, not just the happy path
- Always include GitHub API in find search results (was behind
--federated flag, now default)
- Pass all user-controlled values via env vars instead of
interpolating ${{ outputs.* }} into inline scripts
- Fix operator precedence bug: entry.source || (owner && repo ? ...)
- Use --body-file instead of heredocs with interpolated values
- Quote all variable references in shell commands
Add .github/workflows/auto-publish.yml that triggers on skill-submission
labeled issues, extracts JSON from issue body, validates the source repo
exists, runs security scanner and quality evaluation, then auto-merges
passing submissions (score >= 75) or flags them for manual review.
Add NousResearch/hermes-agent to marketplace/sources.json as a federated
search source for Hermes agent skills.
publish submit:
- Runs quality eval + security scan before submission
- Shows quality badge, stars, last push date, and warnings
- Creates issue directly via gh CLI (no browser needed)
- Falls back to manual URL if gh is not installed/authenticated
- Issue body includes quality score, stars, and push date
check:
- Fetches pushed_at from GitHub API for real update detection
- Shows quality badges next to skill names
- Shows stars and push date in verbose mode
- Displays average quality score across all skills
- findSkill: fall through to .md check when parseSkill returns null
- Use isPathInside instead of dir.includes(cwd) for location detection
- Optional chaining on meta.source in remove --source filter
- Track fence delimiter type per CommonMark (backtick vs tilde)
- Wrap statSync in try-catch so one bad source doesn't abort all installs
- Add cleanup fallback when sparse-checkout/checkout fails
Standalone .md skills (e.g. doc_explanation.md) now get metadata saved
as .doc_explanation.skillkit.json in the same directory. This allows
remove --source to find and remove them. Updated saveSkillMetadata,
loadSkillMetadata, loadMetadata, and remove command to handle both
directory and sibling metadata patterns.
- skillkit remove --all: removes all installed skills
- skillkit remove --source iii-hq/iii: removes all skills from that source
- Both check .skillkit.json metadata to match source
getSearchDirs now collects skill directories from ALL agent adapters
instead of only the configured agent. This fixes 'skill not found' when
skills were installed to .claude/skills/ but the default agent is
universal (which only searches skills/).
findSkill now also checks name.md for standalone .md skills that were
installed as plain files.
Move isStandaloneFile detection and .md suffix normalization before
the existence/force/symlink checks so all code paths use the same
effective target path. Fixes --force not detecting previously installed
standalone skills and symlinks pointing to the wrong filename.
- Restrict try/catch to only the partial clone spawn so
sparse-checkout/checkout failures bubble up as real errors instead
of being misreported as "partial clone unsupported"
- Cap stderr accumulator to 4KB to bound memory on verbose clones
- Add -- terminator before sparsePaths in sparse-checkout set to
prevent user-controlled subpath from being parsed as git options
Instead of wrapping standalone .md skills in a directory, copy them
as-is with .md extension. Skip saveSkillMetadata for standalone files
since it requires a directory. This preserves the original format that
repos like iii-hq/iii use for lightweight single-file skills.
Standalone .md skills (e.g. doc_explanation.md) were copied as plain
files, then saveSkillMetadata tried to write .skillkit.json inside
them causing ENOTDIR. Now detects file sources via statSync and
creates a directory with SKILL.md inside instead.
- spawnGit: set stdout to 'ignore' instead of 'pipe' to prevent
buffer deadlock (git clone writes to stderr, not stdout)
- tryPartialClone: emit fallback message via onProgress instead of
silently swallowing the error
- scanner: preserve fence markers so PI011 delimiter injection is
still detected, only strip inner code block content
- scanner: guard against unclosed frontmatter by looking ahead for
closing --- before entering frontmatter mode
cloneRepo now accepts a subpath option. When set, sparse-checkout
includes both the subpath itself and SKILL_DISCOVERY_PATHS under it.
All four providers pass their parsed subpath through to cloneRepo.
Without this, sparse checkout would skip the subpath directory and
discoverSkills would find nothing.
Strip fenced code blocks and standard YAML frontmatter before scanning
markdown files. This eliminates three categories of false positives:
- PI013: no longer flags standard --- frontmatter delimiters
- CI007: no longer flags curl|sh in code examples
- DE003: no longer flags process.env in code examples
Real threats in prose (outside code blocks) are still caught.
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.
The TUI had a manually-maintained 44-agent AgentType union in
translator.service.ts that fell out of sync when openclaw was added
to core. Replace with re-export from @skillkit/core so it stays in
sync automatically. Fixes CI typecheck failure from #86.
* 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
- Remove legacy clawdbot path checks from OpenClawAdapter.isDetected()
so pure clawdbot workspaces are not hijacked by OpenClaw detection
- Add '.openclaw/agents' to ALL_AGENT_DISCOVERY_PATHS so agent
discovery works for OpenClaw agent directories
* 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
Ran `npx tessl skill review --optimize` on all methodology pack
skills and find-skills. Key improvements:
- Better descriptions with "Use when..." clauses and concrete verbs
- Lowercase-hyphenated name format for consistency
- Trimmed self-evident prose Claude already understands
- Added executable examples and quick reference tables
- Condensed verbose sections into compact tables/summaries
- Fixed YAML parsing errors (unquoted colons in descriptions)
Net result: -552 lines across 14 files.
* feat: add Context Engineering and Antigravity Awesome Skills sources
Add two major community skill repositories to the marketplace:
- muratcankoylan/Agent-Skills-for-Context-Engineering (12.6K stars):
13 skills on context engineering, cited in Peking University paper
- sickn33/antigravity-awesome-skills (16.8K stars):
954+ agentic skills across 9 categories for Claude Code, Gemini CLI,
Cursor, Copilot and more
Sources: 31 → 33 curated collections
Skills: 15,079 → 15,136 indexed entries
* fix: deduplicate skill IDs and fix slack/notion ID mismatch
- Fix ID mismatch: slack-automation → notion-automation (name/desc were Notion)
- Remove 16 generic duplicate entries that clashed with detailed ones
- Total skills: 15,136 → 15,120 after dedup
* fix(extension): replace server API with local content extraction
Chrome Web Store rejected the extension because the core "Save page as
Skill" feature called https://skillkit.sh/api/save-skill which is
unreachable during review. This rewires page saving to work entirely
locally using chrome.scripting.executeScript to extract page content
(title, meta description, headings, body text) directly in the browser.
- Remove API_URL and callSaveApi() server dependency
- Add extractPageContent() via chrome.scripting.executeScript
- Add detectTags() with weighted keyword scoring (URL/headings/body)
- Add buildPageSkill() to generate SKILL.md with frontmatter locally
- Add scripting permission to manifest
- Pass tabId from popup to background for script injection targeting
- Block chrome-extension:// and chromewebstore.google.com URLs
- Add console.error logging in catch blocks for debuggability
- Guard currentTabId validity in popup click handler
- Update store listing to reflect fully local processing
- Bump version to 1.20.0
* fix(extension): address code review findings for robustness
- Extract RESTRICTED_PREFIXES and isRestricted() into shared constants.ts
so popup and background use the same URL blocklist
- Fix yamlEscape to quote YAML bare scalars (true/false/null/yes/no/on/off)
and numeric strings that parsers would coerce to non-string types
- Apply yamlEscape to skill name fields in both buildPageSkill and
buildSelectionSkill to prevent YAML misinterpretation
- Switch tag detection from substring matching to word-boundary regex
to eliminate false positives (e.g. "js" matching "adjustments")
- Check chrome.runtime.lastError in download callback and always
revoke blob URL to prevent leaks on download failure
* 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.
* 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.
Remove WhySkillKit, StackBuilder, BadgeGenerator, Commands, and
TeamEnterprise sections to streamline the landing page from 14 to 9
sections. Replace Extension nav link with Docs, remove duplicate Docs
link, and add a lightweight docs CTA between AdvancedFeatures and
UseCases.
Wire up 3 new screens from PR #65's session intelligence commands into the TUI with
keyboard navigation (l=timeline, v=lineage, f=handoff) and a new Session sidebar section.
* feat: redesign website with aggregator positioning and competitor landscape
- Add WhySkillKit section with competitor comparison table (Vercel skills, PRPM, Askill, Skild, Paks)
- Show SkillKit as meta-package manager aggregating 31+ sources, not competing with them
- Add source grid showing all curated repositories (Anthropic, Vercel, Expo, Supabase, Stripe, etc.)
- Update Hero tagline: "One CLI. Every Source. Every Agent. Zero Telemetry."
- Fix agent count from 32 to 44 in Features stats
- Reframe comparison table: "Simple Installers" vs "SkillKit" instead of vague "Without"
- Update CompatibilityMatrix with all 44 agents (was 16)
- Add zero-telemetry, Apache 2.0, no-account trust signals
- Update nav to include "Why SkillKit" link
* feat: update terminal demo to showcase aggregator workflow
Show install from multiple sources (Anthropic, Vercel), auto-translation
to 44 formats, security scanning, and cross-source recommendations.
* feat: update promo video with aggregator positioning
Updated Remotion video to reflect SkillKit's meta-package manager identity:
- Version bumped from v1.7.5 to v1.19.0 across all scenes
- Install scene shows multi-source aggregation + security scan
- Find scene searches 31+ sources (15K+ skills)
- Translate scene shows cross-source bulk translation
- Workflow scene replaced with security scanner (46 rules)
- Recommend scene shows cross-source recommendations with source labels
- TUI scene updated with source attribution and 44 agents
- Manifest scene shows multi-source team sync
- Logo scene: "One CLI. Every Source. Every Agent." + "Zero Telemetry"
- Agent count updated from "17 agents" to "44 agents | 31+ sources | 15K+ skills"
* feat: move extension save-skill to server-side API
Chrome Web Store rejected the content script approach since it doesn't
inject on pre-existing tabs. Move webpage-to-skill conversion from
client-side to a server-side API route that uses Turndown + weighted
tag detection, eliminating the need for content scripts entirely.
- Add POST /api/save-skill route with SSRF protection and rate limiting
- Rewrite extension to call API with tab URL instead of injecting scripts
- Remove content script, scripting permission, and turndown dependency
- Add name field validation and YAML-escape source URLs
- Bump extension manifest to v1.18.0
* fix: update privacy policy and store listing for server-side save
Privacy policy previously stated "no external requests" which is now
inaccurate. Updated to clearly disclose that the page URL is sent to
agenstskills.com for processing, what data is NOT sent, and that
nothing is stored server-side.
Store listing updated to match: removed "no server" claims, added
data flow explanation, permissions justifications, and data usage
disclosure section for the Chrome Web Store Privacy tab.
* fix: address Devin and CodeRabbit review findings
- Fix SSRF IPv6 checks blocking legit domains (ffmpeg.org, fdroid.org)
by requiring colon in fc00:/ff prefixes (Devin)
- Add 5MB response body size limit to prevent OOM on large pages,
applied to both extractFromUrl and fetchGitHubContent (CodeRabbit)
- Guard response.json() in callSaveApi against non-JSON error pages
from proxies returning HTML 502/504 (CodeRabbit)
- YAML-escape source URL in buildSelectionSkill to match server-side
handling of URLs with #, :, or query strings (CodeRabbit)
* 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
* fix: MCP server fails to start via npx — symlink + skills.json issues
1. isMain guard: process.argv[1] (symlink path) never matched
import.meta.url (real path). Fixed with realpathSync comparison.
2. skills.json not in npm package: prebuild copies marketplace/skills.json
to data/ dir, included in package files.
3. Version string bumped from 1.12.0 to 1.16.0.
Closes#58
* fix: guard realpathSync against undefined argv and resolve both paths
- Wrap in try/catch so library imports don't crash
- Check process.argv[1] exists before calling realpathSync
- Also resolve import.meta.url with realpathSync for pnpm/hoisted setups
* fix: remove redundant GitHub Star button from Hero section
Star count is already visible in navbar and stats bar on all screen sizes.
* fix: keep GitHub Star button on mobile, hide on desktop
Star button uses sm:hidden so it only appears on mobile where
the navbar star badge is less prominent. On desktop (sm+), the
navbar star badge is sufficient.
- Website Hero: "Add to Chrome" button (white, high contrast) between
install command and GitHub star button
- Website navbar: "Extension" link in desktop nav
- Fumadocs: new chrome-extension.mdx page with install, usage,
architecture, permissions, and build docs
- README: Chrome extension section + link in header nav
- Extension README: setup and usage guide
Fully browser-based — no local server required. Content script extracts
page content as markdown (Turndown), background generates SKILL.md with
frontmatter and auto-tags, and downloads via chrome.downloads API.
- Manifest V3 with contextMenus, activeTab, downloads permissions
- Content script: HTML-to-markdown via Turndown, extracts article/main
- Background: SKILL.md generation (slugify, YAML frontmatter, tag detection)
- Downloads to skillkit-skills/{name}/SKILL.md in browser Downloads folder
- Monochromatic popup UI matching website design system
- Context menu: "Save page as Skill" / "Save selection as Skill"
- POST /save API endpoint still available for CLI users
* 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.
* style: update README banner and header for 44 agents
* style: add rotating 3D model banner GIF for README
* style: revert to monochromatic SVG banner, update to 44 agents
* 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
Replace fabricated compatibility data with ratings based on actual
SkillKit translator capabilities, agent constraints, and compatibility
scoring from the codebase.
Key changes:
- Categories now reflect real SkillKit features (Translation, Hooks,
MCP & Tools, Context Window) alongside skill categories
- Cline promoted to Tier 1 (full MCP/tools/hooks support)
- Only Claude, Cline, Roo show full MCP & Tools support (accurate)
- Codex/Copilot show partial for most categories (8K context limit)
- No agent claims 100% - scores reflect real translation capabilities
- Removed language-specific categories (Go, Python, TypeScript) that
don't meaningfully differ across agents since skills are text-based
Replace stale hardcoded fallback values (66 stars, 2.4k downloads, v1.9.0)
with empty/zero defaults that show a dash placeholder while loading. Real
values populate once the API responds.
* 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: add live GitHub stars CTA and redesign README banner
- Add live star count to navbar with star icon and bordered button
- Add GitHub Star CTA button in hero with live count from GitHub API
- Update description to "open source package manager"
- Replace plain ASCII README header with SVG banner
- Fix video visibility with clickable thumbnail fallback
- Add GitHub stars badge to README badges row
* fix: reduce gap between install and star buttons
* fix: remove broken placeholder URL from README video section
* 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()
Reduced from 777 lines to 300 lines. Same information, clearer structure:
problem/solution opening, progressive examples, one command reference
section linking to full docs, compact agent table and attribution list.
* docs: rewrite documentation with clean, concise style
Rewrote all core docs pages (index, installation, quickstart, rest-api,
mcp-server, python-client) with problem/solution framing, progressive
code examples, and conversational tone. Updated OpenAPI descriptions
and api.html meta tags to match. 366 additions vs 722 deletions.
* docs(mcp): remove stale skill count, add list_categories tool name
Replace hard-coded "15,000+" with non-specific wording and explicitly
name the list_categories tool in the Browse categories section.
- Create public/api.html with Swagger UI dark theme matching site design
- Create public/openapi.json with full OpenAPI 3.1 spec and examples
- Add Vercel rewrite: /api -> /api.html
- Add "API" nav link to website navbar
- Custom dark theme with monospace fonts matching SkillKit branding
- 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.
Add beautiful-mermaid to the skills marketplace:
- Render Mermaid diagrams as beautiful SVGs or ASCII art
- 5 diagram types: flowchart, sequence, class, ER, state
- 15 built-in themes + Shiki/VS Code theme integration
- Zero DOM dependencies, works in Node.js/Bun/Deno/browsers
Source: rohitg00/beautiful-mermaid
* feat(website): add dynamic stats from npm and GitHub APIs
- Create useStats hook to fetch real-time version, downloads, and stars
- Replace hardcoded values in App.tsx stats bar with dynamic data
- Replace hardcoded version in Hero.tsx badge with dynamic data
- Add 5-minute cache to reduce API calls
- Fallback to default values if APIs fail
* fix: address CodeRabbit and Devin review comments
- Fix duplicate API calls: remove useStats from Hero, pass version as prop from App
- Fix falsy checks: use typeof checks for downloads and stargazers_count
to correctly handle zero values instead of showing defaults
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
* refactor(tui): migrate to OpenTUI/Solid.js framework
Complete TUI rewrite from Ink/React to OpenTUI/Solid.js:
- Replace React with Solid.js for reactive state management
- Migrate all 20 screens to OpenTUI component primitives
- Add monochromatic design with 14 color themes
- Implement cinematic splash screen with ASCII animation
- Add vi-key navigation (j/k/h/l) throughout
- Create 32 reusable components with consistent styling
- Add service layer for @skillkit/core API integration
- Remove deprecated react-opentui.d.ts type definitions
- Update build config for Solid.js JSX transform
Features covered:
- Browse, Marketplace, Recommend (discovery)
- Installed, Sync, Translate (management)
- Workflow, Execute, Plan, History (execution)
- Team, Context, Memory, Mesh, Message (collaboration)
- Plugins, Methodology, Settings (configuration)
- Help screen with 25+ keyboard shortcuts
* fix(tui): address CodeRabbit review feedback
Critical fixes:
- Remove non-existent module exports from index.tsx (context, themes, ui, hooks, services)
- Simplify App.tsx to remove missing context provider imports
Major fixes:
- ProgressBar: Add defensive guards for width/progress (NaN, Infinity, negative)
- Browse: Fix filtered index loading wrong repository
- Context: Add section navigation (h/l keys) and async error handling
- Methodology: Add error handling with try/catch/finally for all async ops
- Plan: Fix division by zero when plan has no tasks, add error handling
- Team: Add error handling for loadTeamConfig and handleInitialize
- Translate: Guard against missing skill.path before read operations
* fix(tui): address remaining CodeRabbit feedback
- Memory: Add error handling for loadData, handleDelete, handleClearAll, handleInitialize
- Team: Fix key conflict by blocking global keys when detail pane is open
- animations: Fix springCurve to return 0 at progress=0 and 1 at progress=1
* fix(tui): add Solid.js OpenTUI type definitions
Required for TypeScript to recognize OpenTUI JSX elements (box, text, span)
with their custom props (fg, bg, flexDirection, etc.)
* fix(tui): add stub services and fix type errors for CI
- Add minimal stub service implementations for all TUI screens:
- context.service.ts - project context loading
- executor.service.ts - skill execution
- memory.service.ts - memory management
- methodology.service.ts - methodology packs
- plan.service.ts - plan management
- plugin.service.ts - plugin management
- recommend.service.ts - skill recommendations
- team.service.ts - team collaboration
- translator.service.ts - skill translation
- workflow.service.ts - workflow orchestration
- Fix type errors in screens:
- Plan.tsx: Add non-null assertions for optional issues array
- Plugins.tsx: Cast getPluginInfo result to Plugin type
These stubs provide the expected interfaces for screens while
the full integration with @skillkit/core is completed later.
* fix(tui): add missing component files for CI
Add the component files that screens depend on:
- StatusIndicator - loading/success/error status display
- DetailPane - side panel for detailed view
- EmptyState/ErrorState - empty and error state components
- SelectList - selection list component
- SplitPane - split view layout
- AnimatedText - text animations
- Breadcrumb - navigation breadcrumb
- Button/Clickable - interactive elements
- CodeBlock - code display
- FormField - form input wrapper
- HoverHighlight - hover effect wrapper
- RightSidebar/BottomStatusBar - layout components
- TabBar - tab navigation
- ErrorBoundary - error handling wrapper
These components are required by the screen files and were
missing from the previous commit.
* fix(tui): resolve context import and duplicate export issues
- RightSidebar: inline sidebar state instead of importing from context
- services/index.ts: fix duplicate AgentType export by using explicit
exports for executor and translator services
* fix(tui): use export type for type-only re-exports
* refactor(tui): simplify components and remove redundant exports
* refactor(tui): remove unused imports and variables
Code simplification pass:
- Remove unused onCleanup import from Workflow, Recommend, Sync screens
- Remove unused contentWidth from Browse, Recommend, Marketplace, Sync
- Remove unused cols from Browse, Recommend
- Remove unused JSX import from EmptyState component
- Remove unused translate import from Sync screen
- Remove unused SelectList import and listItems function from Memory screen
* fix(tui): wire up interactive handlers and fix type definitions
- AnimatedText: fix reactive dependency tracking in createEffect
- BottomStatusBar: remove unused width prop
- Breadcrumb: wire onNavigate handlers, fix off-by-one in getPathAtIndex
- Button: wire onClick with disabled check, apply paddingX, fix ButtonGroup spacing
- Clickable: wire click/double-click/right-click handlers, fix InteractiveArea
- DetailPane: make close button interactive
- ErrorBoundary: use terminal keyboard handler, prevent duplicate onError calls
- FormField: fix placeholder show condition for empty string
- SelectList: add onSelect/onHover interactive behavior
- SplitPane: preserve string sizes like "50%" in width/height helpers
- TabBar: wire onSelect callbacks for both horizontal and vertical variants
- solid-opentui.d.ts: add missing event handler types to BoxProps/TextProps
* fix(tui): address Devin review issues
- App: restore process.exit(0) fallback when onExit is undefined
- Execute: pass skill path instead of content to executeSkillWithAgent
- Memory: add useKeyboard handler for j/k/Enter/d/c/i/Esc shortcuts
* fix(tui): address Devin and CodeRabbit review issues
Security & Robustness:
- App: use execFile instead of exec to prevent command injection in openUrl
- App: clamp cols/rows to minimum 1 to prevent negative dimensions
- index: await render() and handle failures with exitTUI(1)
Execute screen:
- Add error state for load failures
- Prevent re-entrant execution with executing() guard
- Use try/finally to always clear executing state
Memory screen:
- Add missing 'r' shortcut for retry/refresh
Workflow screen:
- Use try/finally to always clear executing state even on rejection
Sync screen:
- Capture selectedIndex and agent values before await to prevent drift
* fix(tui): implement list windowing to keep selection visible
Execute, Marketplace, Recommend screens now use windowed list rendering
that follows the selection, preventing the selected item from being
off-screen when navigating through long lists.
- Add createMemo for computing visible window based on selectedIndex
- Show ▲/▼ indicators when items are above/below the visible window
- Navigate through full list instead of clamping to visible window size
* fix(tui): add error handling and index clamping
Marketplace.tsx:
- Set error signal when all repos fail or no skills loaded
- Add reactive effect to clamp selectedIndex when filteredSkills changes
Recommend.tsx:
- Wrap loadData in try/catch to surface errors in UI
- Clamp selectedIndex after recommendations update
* fix(tui): add error handling to Plugins and Workflow screens
Plugins.tsx:
- Wrap loadData in try/catch to prevent stuck loading state
- Wrap handleToggle in try/catch/finally to always reset toggling flag
- Wrap handleShowDetail in try/catch to surface errors gracefully
Workflow.tsx:
- Wrap loadData in try/catch to prevent stuck loading state
- Add catch block to handleExecute to show execution errors in UI
* feat(website): add demo video to Hero section
- Add skillkit.mp4 video player below hero content
- Auto-plays muted with loop for seamless demo experience
- Styled with terminal-like window frame
* fix(website): fix Twitter/OG preview card
- Update og:url from skillkit.dev to agenstskills.com
- Use absolute URLs for og:image and twitter:image
- Add og:image dimensions (1200x630)
- Add og:site_name and twitter:url
- Update description to "32 agents"
* feat(docs): comprehensive documentation and website improvements
Website improvements:
- Add AdvancedFeatures component for Memory, Primer, Mesh, Messaging
- Add UseCases component with real-world scenarios
- Add TeamEnterprise component for collaboration features
- Update Hero section with better value proposition
- Make all components mobile responsive
- Pre-populate StackBuilder with default skills
- Add ARIA attributes for accessibility
- Add prefers-reduced-motion for smooth scrolling
Documentation updates:
- Create primer.mdx for auto-generate instructions feature
- Create teams.mdx for team collaboration documentation
- Enhance index.mdx with comprehensive feature overview
- Update quickstart.mdx with advanced features section
- Expand api-reference.mdx with Mesh, Messaging, Workflow APIs
- Update configuration.mdx with Memory, Mesh, Primer settings
- Simplify fumadocs homepage for cleaner presentation
- Fix markdown formatting issues
* fix: address CodeRabbit review suggestions
- Add missing fs import to api-reference.mdx code examples
- Clarify that host comes from MeshHost in messaging example
- Remove broken /docs/registries link from teams.mdx
- Fix ARIA tab/panel linkage in AdvancedFeatures.tsx
* feat(docs): comprehensive documentation and website improvements
Website improvements:
- Add AdvancedFeatures component for Memory, Primer, Mesh, Messaging
- Add UseCases component with real-world scenarios
- Add TeamEnterprise component for collaboration features
- Update Hero section with better value proposition
- Update Features section with improved presentation
- Make all components mobile responsive
- Pre-populate StackBuilder with default skills
- Hide Apache 2.0 on mobile, show Docs link on mobile
Documentation updates:
- Create primer.mdx for auto-generate instructions feature
- Create teams.mdx for team collaboration documentation
- Enhance index.mdx with comprehensive feature overview
- Update quickstart.mdx with advanced features section
- Expand api-reference.mdx with Mesh, Messaging, Workflow APIs
- Update configuration.mdx with Memory, Mesh, Primer settings
- Simplify fumadocs homepage for cleaner presentation
* fix: address code review feedback
- Fix markdown heading/list separation in index.mdx
- Fix mesh.security schema in configuration.mdx profiles
- Add shebang to git hook script in primer.mdx
- Update Primer description for multi-agent support in Features.tsx
- Add prefers-reduced-motion media query for accessibility in styles.css
- Add ARIA attributes (role, aria-selected, aria-controls) to AdvancedFeatures tabs
- Add scripts/bump-version.sh for syncing all package versions
- Add version-check job to CI workflow
- Add tag/version validation to publish workflow
- Sync all packages to v1.8.0
- Add VERSIONING.md documentation
- Add Mesh screen for mesh network management
- Add Message screen for inter-agent messaging
- Update App.tsx with navigation to new screens
- Add 'g' and 'j' keyboard shortcuts for mesh/message
- Update README with mesh and message CLI commands
- Update version from v1.7.5 to v1.7.10 in Hero
- Add primer command to terminal demo animation
- Add primer to core capabilities in Documentation
- Add AI Instruction Generation (Primer) section to CLI Commands
- Add Primer feature card to Features section
- Add Instructions comparison row highlighting Primer
Fixed getAllAdapters() to return the correct agent type from the
record keys instead of relying on adapter.type property, which
was returning 'universal' for all UniversalAdapter instances.
This fixes the E2E test that checks for all 32 agents in init --list
output (cline, codebuddy, commandcode, etc. were showing as 'universal').
* 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
- Add docs badge and link to README
- Add all new commands: check, find, manifest, fix, audit, publish, status, pause/resume, settings, methodology, plan, plugin, team, hook, command
- Reorganize commands into logical sections
- Update TUI Help screen with new shortcuts
- Add 'd' key to open docs in browser from TUI
- Add noopener to rel="noreferrer" for tabnabbing protection
- Fix standalone skill path handling in readSkillContent and validateSkill
- Derive TOTAL_AGENTS from adapters instead of hardcoded value
- Fix duplicate discovery path for droid/factory agents
- Create dedicated FactoryAdapter instead of using UniversalAdapter
- Show Docs link on mobile (removed hidden sm:block)
- Add proper overflow handling for all code blocks on mobile
- Use smaller padding/text sizes on mobile screens
- Add break-words and whitespace-pre-wrap to prevent overflow
* docs: add comprehensive documentation section to website
- Create Documentation.tsx component with DeepWiki-sourced content
- Add sections: Overview, Installation, Architecture, Agents, Commands
- Add sections: Configuration, API Reference, Skill Format, Tech Stack
- Include interactive sidebar navigation
- Add Docs link to navbar
- Reference DeepWiki as documentation source
Source: https://deepwiki.com/rohitg00/skillkit
* feat(docs): add Fumadocs documentation site and update website
- Add Fumadocs-based documentation site at /docs
- Complete MDX documentation for all features
- TUI section with keyboard navigation
- Installation, quickstart, commands reference
- API reference and configuration guides
- Update SkillKit website (agentskills.com)
- Add 15 new agents (32 total): CodeBuddy, Continue, Crush,
Factory, MCPJam, Mux, Neovate, OpenHands, Pi, Qoder, Qwen,
Vercel, Zencoder, CommandCode
- Update agent count from 17 to 32 across all components
- Update command count from 30+ to 40+
- Remove inline Documentation section (now in Fumadocs)
- Link Docs to /docs route
- Add 15 new agent adapters to @skillkit/agents
- All use UniversalAdapter as base
- Added to detection order and adapter registry
- Configure Vercel rewrites for /docs proxy to Fumadocs
* fix: resolve TypeScript build errors in fumadocs
- Remove createRelativeLink from docs page (not compatible with current fumadocs)
- Remove redundant isLoading prop in SkillGenerator (status already narrowed)
- Fix lucide-react icon type casting in Card component
- Fix MDXComponents return type in mdx-components
* fix: resolve /docs 404 and show docs link on mobile
- Handle undefined slug by defaulting to empty array for index page
- Remove hidden class from Docs link so it shows on mobile
* fix: address CodeRabbit review suggestions
- Move Gemini API key to server-side route (security)
- Add default type='button' to Button component
- Use dynamic agent count instead of hardcoded 32
- Add 'use client' directive to Documentation
- Fix stale closure in SkillSubmitForm with useCallback
- Add URL validation for external references
- Add noopener to external links for security
- Fix Clawdbot path in agents.mdx
- Update cicd.mdx with correct CLI commands
- Capitalize proper nouns (Markdown, provider names)
- Add text language to README code block
- Update date example in quickstart.mdx
* fix: add noopener to external links and improve nav accessibility
- Add noopener to rel attributes on external links for tabnabbing protection
- Add aria-label to nav element for screen readers
- Add type="button" and aria-current to nav buttons
* fix: add vercel.json for fumadocs deployment
* fix: disable turbo detection in Vercel deployment
* fix: clean up ASCII logo for better rendering
* fix: improve code block contrast in SkillGenerator
* fix: sync ASCII logo and code block contrast fixes to old Vite components
* fix: improve mobile responsiveness for tabs and header buttons
* fix: address CodeRabbit security and code quality suggestions
- Add rate limiting to API route (10 requests/min per IP)
- Add proper type validation for topic parameter
- Move API key check after input validation
- Add 'use client' directive to SkillGenerator
- Add YAML escape helper to prevent frontmatter injection
- Change vercel.json to use pnpm instead of npm
- Remove deprecated @types/minimatch dependency (minimatch 10.x has bundled types)
- Fix TUI text rendering overlap by using single text elements instead of multiple
text elements in box rows (Browse, Installed, Marketplace, SkillList screens)
- Fix terminal crash on 'q' press by properly unmounting React tree and restoring
terminal state with escape sequences
- Add signal handlers (SIGINT, SIGTERM) for clean exit
- Bump version to 1.7.6
* feat(cli): add beautiful onboarding UI with @clack/prompts
- Add new onboarding module with theme, logo, prompts, summary, and preferences
- Implement adaptive ASCII logos (full/compact/minimal based on terminal width)
- Add agent icons (⟁ ◫ ◎ ✦ ⬡) and circle-based step indicators (● ○)
- Create agentMultiselect and skillMultiselect with visual formatting
- Add progress bars and score formatting utilities
- Store user preferences in ~/.skillkit/preferences.json
- Update all interactive commands with new UI (init, install, sync, marketplace, recommend)
- Add unit tests (18 tests) and e2e tests (16 tests)
- Replace nested ternaries with if/else for clarity
* fix(cli): address review feedback from CodeRabbit and Devin
- Fix symlink install method: actually create symlinks for subsequent agents
- Fix validate type signature to match @clack/prompts API (string | Error | undefined)
- Handle zero detected agents to avoid empty multiselect blocking
- Clamp divider width to avoid negative repeat counts
- Guard progressBar against invalid totals/values
- Isolate preference tests from real ~/.skillkit state
* fix(cli): TypeScript type compatibility with @clack/prompts
- Cast select options to match Option<T> type signature
- Cast groupMultiselect options for proper Record type
- Fix tasks function return type to Promise<string | void>
* fix(cli): use 'as any' for @clack/prompts type compatibility
The Option<T> type from @clack/prompts has a complex structure that
doesn't match our inline types. Using 'as any' assertion to bypass
strict type checking while maintaining runtime correctness.
* feat(ci): add GitHub Packages publishing
Adds a new job to publish the main skillkit package to GitHub Packages
(npm.pkg.github.com) as @rohitg00/skillkit after successful npm publish.
- Publish to npm first, then GitHub Packages
- Uses GITHUB_TOKEN for authentication (no additional secrets needed)
- Package available at npm.pkg.github.com/@rohitg00/skillkit
* docs: add GitHub Packages installation instructions
* fix(ci): resolve workspace dependencies before GitHub Packages publish
* feat: single source of truth for skills marketplace
- Consolidate all skills data to marketplace/skills.json (15,062 skills)
- Add marketplace/sources.json for curated repository sources (31 repos)
- Add skillkit publish command for user skill submissions
- Add auto-publish prompt to skillkit init
- Add AI skill caching with localStorage (7-day TTL)
- Add Submit to Marketplace button for AI-generated skills
- Remove duplicate skills.json files (docs, core)
- Remove legacy src/tui/ folder (replaced by packages/tui)
- Update all imports to use single source files
* fix: address CodeRabbit review feedback
- Remove duplicate checklist items in SkillSubmitForm.tsx
- Add missing sources-schema.json file
- Add slugify method for URL-safe skill IDs
- Fix Windows browser open command (empty title param)
- Fix frontmatter parsing for CRLF and multiline tags
* fix: address additional review feedback
- Add TTY check to skip prompts in CI/non-interactive environments
- Guard against empty slug IDs with validation
- Print full URL in fallback (not truncated)
- Fix repo name parsing for names containing dots
- Add cache pruning and max entries limit (50) for localStorage
* fix: filter empty strings from YAML tags array parsing
- Consolidate all skills data to marketplace/skills.json (15,062 skills)
- Add marketplace/sources.json for curated repository sources (31 repos)
- Add skillkit publish command for user skill submissions
- Add auto-publish prompt to skillkit init
- Add AI skill caching with localStorage (7-day TTL)
- Add Submit to Marketplace button for AI-generated skills
- Remove duplicate skills.json files (docs, core)
- Remove legacy src/tui/ folder (replaced by packages/tui)
- Update all imports to use single source files
- Add og-image.png for social media previews
- Add Open Graph meta tags (title, description, image, url)
- Add Twitter Card meta tags (summary_large_image)
- Add meta description for SEO
- Hero: use items-start for proper top alignment
- Agents: add pl-8 padding to prevent first icon cutoff
- Update all npx commands to use skillkit@latest
- Validation errors now stack vertically for better readability
- Footer more compact with smaller padding
- Shorter footer text on mobile
- Better responsive breakpoints (sm instead of md)
- Use inline-flex for tight button sizing
- Remove unnecessary wrapper div and gradient
- Use flex-wrap for responsive layout
- Reduce spacing for cleaner look
- Hide terminal animation on mobile (only show on lg screens)
- Add 'Copied!' text feedback when clicking npx command
- Make npx button full width on mobile
- Reduce blank space on mobile hero section
- Add max-w-5xl constraint to Agents scrolling section
- All major sections now have consistent max-width
- Reduces stretched appearance on large monitors
- Change from anchor tag to button with explicit window.open
- Ensures click handler properly opens GitHub issue URL
- Add disabled state to button when form is invalid
- Add max-w-5xl mx-auto to Hero grid container
- Add max-w-5xl mx-auto to Features comparison and feature cards grids
- Set fixed terminal width lg:max-w-[420px] for consistent sizing
- Reduce gap on mobile for better spacing
* feat(marketplace): add community skill submission system
- Add marketplace/skills.json with curated skills collection
- Add marketplace/schema.json for skill validation
- Add marketplace/CONTRIBUTING.md with submission guidelines
- Add skill submission issue template for community contributions
- Add workflow to auto-process approved skill submissions
- Add bundled skills data in core package for offline access
Community members can now submit skills via GitHub Issues, which are
automatically added to the marketplace when approved by maintainers.
* fix: address CodeRabbit review suggestions for marketplace PR
Workflow improvements:
- Add concurrency block to prevent parallel push conflicts
- Fix authorGithub extraction to preserve hyphens in usernames
- Filter empty tags and validate at least one tag exists
Schema improvements:
- Allow dots in repository names (e.g., user/repo.js)
- Copy schema.json to core package data folder for validation
Documentation:
- Add 'text' language tag to code block in CONTRIBUTING.md
Data sync:
- Add skills.json and schema.json to docs/skillkit/data/
* refactor(docs): simplify website components and update README
Website Code Simplification:
- Convert arrow functions to named function declarations with explicit return types
- Extract inline styles to module-level constants (BASE_STYLES, VARIANT_STYLES, etc.)
- Extract SVG icons to constants (GITHUB_ICON, CHECK_ICON, COPY_ICON)
- Extract CSS animations to constants (FADE_ANIMATION_STYLES, SCROLL_ANIMATION_STYLES)
- Add explicit TypeScript interfaces (Feature, Agent, TerminalLine, etc.)
- Replace nested ternaries with switch statements for clarity
- Extract inline handlers to named functions (scrollToTop, openGitHub, etc.)
- Improve variable naming (idx -> index, cap -> feature)
- Use meaningful React keys instead of index-only keys
README Updates:
- Remove "What's New in v1.6" section (release notes should be in CHANGELOG)
Components updated:
- App.tsx, Button.tsx, Card.tsx, Terminal.tsx, Footer.tsx
- Features.tsx, Agents.tsx, Hero.tsx
- SkillSubmitForm.tsx, SkillGenerator.tsx
* fix: address CodeRabbit review suggestions
Security:
- Add noopener,noreferrer to window.open in Hero.tsx
- Handle clipboard API errors with try/catch
Accessibility:
- Add Space key support to Card.tsx for keyboard navigation
- Add aria-label to GitHub icon link in Footer.tsx
Error Handling:
- Fix useCallback dependency for processMarkdownFile in SkillSubmitForm
- Add input validation for topic in geminiService.ts
Data Validation:
- Add defensive checks for malformed source URLs in skills-index.ts
- Generate proper skill ID for upload mode in SkillSubmitForm
Type Safety:
- Replace unsafe 'as any' cast with proper type annotation in geminiService
Code Quality:
- Use array filter pattern for class concatenation in Button.tsx
* feat: update agent adapters for 2026 compatibility
- Add new skill-translator module for cross-agent skill translation
- Update Cursor adapter to use .cursor/rules/*.mdc format with YAML frontmatter
- Update Windsurf adapter to use .windsurf/rules/*.md with mode field
- Update GitHub Copilot to use .github/copilot-instructions.md
- Update Trae to use .trae/rules/project_rules.md with YAML frontmatter
- Update OpenCode global path to ~/.config/opencode/
- Add new discovery paths for Cursor, Kilo, Roo, Windsurf, GitHub Copilot
- Improve agent detection with additional config file checks
BREAKING CHANGE: Config file paths have changed for Cursor, Windsurf, Trae, and GitHub Copilot adapters to match their 2026 specifications.
* feat: add e2e tests for skill-translator and fix translation issues
- Add comprehensive e2e tests for skill translator (86 tests)
- Fix translateSkillToAll to include all 17 agents
- Fix writeTranslatedSkill to check for failed translations
- Update vitest config to resolve workspace packages
- Remove unnecessary 2026 comments from agent adapters and skill-translator
* fix: improve agent detection and add overwrite option to translator
Agent detection fixes:
- github-copilot: Remove overly broad .github dir check, only detect
Copilot-specific artifacts (copilot-instructions.md, instructions, ~/.copilot)
- kilo: Fix modes path from .kilocodemodes to .kilocode/modes
- roo: Fix modes path from .roomodes to .roo/modes
Translator improvements:
- Add overwrite option to writeTranslatedSkill() to prevent accidental overwrites
- Return skipped flag when file exists and overwrite=false
Refactoring:
- Centralize agent config in packages/core/src/agent-config.ts
- Update all agent adapters to use shared AGENT_CONFIG
- Re-export utility functions for backwards compatibility
* fix: update clawdbot test and use workspace dependency
- Update clawdbot test to expect new standardized skillsDir (.clawdbot/skills)
- Change @skillkit/core dependency to workspace:* for proper monorepo linking
* fix: address CodeRabbit and Devin review suggestions
- Remove AGENTS.md from detection in Kilo, Roo, Trae adapters (prevents false positives)
- Fix AmpAdapter detection path (check .amp instead of .agents)
- Fix regex pattern in parseConfig for markdown adapters to match any skill name format
- Add hooks and permissionMode to specialFields for proper extraction
- Add path traversal protection in writeTranslatedSkill
- Fix boolean field preservation using 'in' check instead of truthy checks
- Add mode field preservation for Windsurf
- Add permissionMode support for Roo adapter
- Add --recursive flag to scan nested directories for agents
- Add --source flag to specify custom source directory
- New functions: discoverAgentsRecursive, discoverAgentsFromPath
- Add 13 new tests for recursive discovery
- Update README with agent commands documentation
- Bump version to 1.6.3
The fallback sample skills were referencing 'skillkit/samples' which
is a non-existent GitHub repository. Changed to 'built-in' to clearly
indicate these are bundled sample skills.
Fixes#9
* feat: add custom AI sub-agents support (Phase 19)
Add full support for custom AI sub-agents (.claude/agents/*.md):
Core module (packages/core/src/agents/):
- types.ts: Agent types, schemas, AGENT_DISCOVERY_PATHS for 17 AI agents
- parser.ts: Parse agent markdown with YAML frontmatter
- discovery.ts: Discover agents across all supported agent directories
- translator.ts: Translate agents between Claude, Cursor, and universal formats
- index.ts: Export all agent functionality
CLI commands (skillkit agent):
- list: List all installed agents (project + global)
- show: Show agent details
- create: Create a new agent from template
- translate: Translate agents between formats
- sync: Sync agents to target AI coding agent
- validate: Validate agent definitions
Features:
- Agent discovery across 17 AI coding agents
- Frontmatter support: name, description, model, permissionMode,
disallowedTools, allowedTools, hooks, skills, context, tags, author, version
- Translation between claude-agent, cursor-agent, and universal formats
- 36 new tests (706 total passing)
Closes#5
* fix: use workspace:* for internal dependencies
pnpm automatically converts workspace:* to actual versions when publishing,
but uses local workspace packages during development and CI.
* fix: replace require() with ESM imports and add name validation
- parser.ts: Import statSync instead of using require('node:fs')
- translator.ts: Add ESM imports for extractAgentFrontmatter, extractAgentContent, AgentFrontmatter
- agent.ts: Add validation for agent name format (lowercase alphanumeric with hyphens)
Addresses review comments from Devin.
- Change tagline from "One CLI. 17 AI Agents. Zero Lock-in." to
"Supercharge Every AI Coding Agent"
- Add "What are AI Agent Skills?" callout explaining skills to newcomers
- Move problem/solution section higher for immediate visibility
- Update npm package description for better discoverability
- Update GitHub repo description
- Remove redundant "Why SkillKit Exists" section at bottom
- Relocate "What's New in v1.6" after comparison table
- Add comprehensive E2E test suite (180 tests)
- Fix ESM compatibility, security improvements in test runner
- Align E2E tests with actual CLI interface
- Add timeout guards for reliable test execution
- Add 31 unit tests for ClawdbotAdapter covering all functionality
- Fix Clawdbot detection to check workspace skills/, global .clawdbot/, and clawdbot.json
- Fix escapeXml helper to handle undefined values gracefully
- Update README to document workspace and global skills directories
- Brings total test count to 670 tests (all passing)
* feat: add team collaboration and plugin system
Team Collaboration (Phase 10):
- TeamManager for skill sharing with remote registries
- Skill bundles (create, export, import)
- Team CLI commands: init, share, import, list, sync, remove, bundle-*
- Team TUI screen with navigation key 'a'
Plugin System (Phase 11):
- PluginManager with lifecycle hooks
- Plugin loader (files, npm packages, directories)
- TranslatorPlugin for custom agent formats
- ProviderPlugin for custom skill sources
- CommandPlugin for custom CLI commands
- Plugin CLI commands: list, install, uninstall, enable, disable, info
- Plugins TUI screen with navigation key 'p'
Additional:
- CI/CD command for template generation
- Settings command improvements
- Updated CLAUDE.md with all 11 phases complete
- 359 tests passing
* fix: address PR review comments from Devin
- Fix skill-executor.ts: handle undefined agentToUse when strategy is 'cli'
Previously, if strategy was 'cli' but no CLI agent was available, the
code would fall through and call executeWithAgent with undefined agent,
causing a runtime error.
- Fix cicd.ts: prevent GitLab CI config overwrite without --force
Previously, if .gitlab-ci.yml existed but didn't contain 'skillkit',
it would be overwritten without checking --force. Now it properly
returns early with a helpful message, consistent with CircleCI behavior.
* fix: address PR review comments from CodeRabbit and Devin
Plugin command fixes:
- Fix double-nested .skillkit path when using --global flag
Settings command fixes:
- Add defaultTimeout to help text for available keys
- Wrap const timeout in block scope for switch case
- Fix --set flag validation to require equals sign
Core package fixes:
- Remove .ts from supported plugin extensions (runtime imports fail)
- Store skill name instead of full path in bundle manifest
- Use ESM import for execSync instead of require()
- Replace custom YAML parser with yaml library for proper nested structure handling
TUI screen fixes:
- Guard against empty plugins array in navigation
- Fix empty object pattern in Settings.tsx
- Surface config load failures in Settings screen
- Add manager.load() before sync/import in Team screen
- Guard against empty skills array in Team navigation
- Clamp selection index after skills update
CI/CD command fixes:
- Properly handle skipped workflows (not reported as success)
- Show different final message when nothing was created
- Wrap readFileSync in try/catch for GitLab CI
* fix: address PR review feedback (round 2)
- Security: Add path traversal protection in bundle import
- Fix Plugins.tsx to load plugins from directory before listing
- Fix plugin.ts to persist plugin files to disk on install/uninstall
- Fix Settings.tsx auto-detect vs universal display inconsistency
- Add null check for registry in team/manager.ts sync()
- Thread TeamConfig.auth through provider clone calls
- Fix GitLab CI rules vs only conflict (use rules instead of only)
- Make loadConfig() accept global parameter for --global flag
- Fix relative path detection in plugins/loader.ts (../plugins)
* fix: address PR review feedback (round 3)
- plugin.ts: Add '../' and '~' to isLocalPath detection, use
loader-recognized names (index.js, plugin.json) for single-file copy
- plugin.ts: Remove unused 'basename' import
- settings.ts: Change defaultTimeout validation from < 0 to <= 0
- settings.ts: Use knownKeys whitelist for unknown key detection
- loader.ts: Wrap statSync in try/catch for scanDirectory
- loader.ts: Validate required fields (name, version) before pushing
- loader.ts: Update regex to allow scoped npm names (@scope/name)
- bundle.ts: Resolve targetDir to absolute path for path containment
- Plugins.tsx: Only ignore ENOENT errors, rethrow others
- Plugins.tsx: Load plugins in togglePlugin before enable/disable
* fix: address PR review issues - round 4
CRITICAL fixes:
- loader.ts: Wrap statSync in try/catch in loadPluginsFromDirectory to handle
permission denied and broken symlinks
- bundle.ts: Block path traversal attacks via skill names in importBundle by
validating skill names don't contain / \ or start with .
MAJOR fixes:
- bundle.ts: Use dirname() instead of join('..') for output directory calculation
MINOR fixes:
- plugin.ts: Expand tilde paths (~) before existsSync for proper file operations
- settings.ts: Show defaultTimeout and marketplaceSources in default output
- bundle.ts: Prevent duplicate skill names in bundle with early error
- bundle.ts: Handle single-file skills properly in detectAgents
* fix: address PR review issues - round 5
CRITICAL fixes:
- plugin.ts: Add path traversal protection in uninstallPlugin with name validation
and resolved path checks
- bundle.ts: Fix parseSkillContent to not drop first file when content starts
with "--- file ---" header
MAJOR fixes:
- plugin.ts: Add isValidPluginName() helper and validate plugin.metadata.name
before using in file paths
- plugin.ts: Preserve .mjs extension for single-file ESM plugins
- bundle.ts: Use cross-platform path containment checks with relative() and sep
instead of hardcoded '/'
- loader.ts: Add support for .mjs files (index.mjs, plugin.mjs)
MINOR fixes:
- plugin.ts: Expand tilde (~) path BEFORE calling loadPlugin
- settings.ts: Use null instead of undefined in JSON output for --get so key
is preserved
* fix: allow scoped npm names in isValidPluginName
Update isValidPluginName to allow scoped npm package names like @scope/name
while still preventing path traversal attacks. Uses the same regex pattern
as loader.ts validatePlugin for consistency.
Valid names: my-plugin, @scope/plugin-name
Invalid names: ../evil, .hidden, name\\path, @../evil/name
* fix: handle more relative path patterns in loadPlugin
- Use isAbsolute() for cross-platform absolute path detection
- Add check for paths containing '/' (excluding @scope packages)
- Simplify pattern to source.startsWith('.') for all dot-prefixed paths
- Now correctly handles: plugins/x.js, ../foo, ./bar, .hidden
- Main README now leads with "One CLI. 17 AI Agents. Zero Lock-in."
- Clear problem/solution format showing SkillKit's value
- Feature comparison table (with vs without SkillKit)
- Organized command reference by category
- Updated all package READMEs with comprehensive docs:
- @skillkit/core: Full API examples for all features
- @skillkit/cli: Complete command reference
- @skillkit/tui: Screen descriptions and keyboard shortcuts
- @skillkit/agents: Agent adapter documentation
Security fixes:
- Remove shell: true from spawn to prevent shell injection (agents.ts)
- Add safe regex test with ReDoS protection (engine.ts)
- Add security warning for verification commands (engine.ts)
Bug fixes:
- Fix global memory index path to use ~/.skillkit/ (compressor.ts, memory-index.ts)
- Add task_complete events and verification to resumeExecution (engine.ts)
- Handle absolute URLs in skill entries (aggregator.ts)
- Implement actual port availability check (runner.ts)
- Remove invalid JUnit report format (templates.ts)
- Add crash guard to workflow list command (workflow/list.ts)
- Fix resume skipping checkpoint handling in engine.ts
- Add paused status guard before completeExecution call
- Fix NaN% progress when totalSteps=0 in Execute.tsx
- Fix effectiveness ?? instead of || to preserve 0% values in compressor.ts
- Clear cache on source change in aggregator.ts addSource/removeSource
New Features:
- Skill Marketplace: Aggregates skills from GitHub repos with caching
- Skill Testing Framework: 18+ assertion types for skill validation
- CI/CD Integration: Templates for GitHub Actions, GitLab CI, CircleCI
- Workflow Orchestration: Wave-based parallel skill execution
- Session Management: Pause/resume support with state persistence
- Agent Execution: CLI spawning for claude/codex/gemini/opencode/goose
New CLI Commands:
- skillkit marketplace - Browse and install from marketplace
- skillkit test - Run skill tests with assertions
- skillkit run - Execute skills with orchestration
- skillkit status - Show session state
- skillkit pause/resume - Session management
- skillkit workflow - Workflow management
New TUI Screens:
- Marketplace browser with search
- History viewer with session details
- Workflow management screen
Bug Fix:
- Fixed dry-run mode in memory compress command (Devin review)
The --dry-run flag now properly prevents storing learnings
* feat: add TUI enhancements and monorepo structure
- Add Translate screen for cross-agent skill translation
- Add Context screen for project context management
- Add Recommend screen with smart skill recommendations
- Fix TUI screen duplication by using fixed height container
- Clean up unnecessary comments following code-simplifier principles
- Refactor to monorepo structure with @skillkit/core, @skillkit/cli,
@skillkit/tui, and @skillkit/agents packages
- Add comprehensive tests for translator, context, and recommend features
- Update license to Apache-2.0
- Add CONTRIBUTING.md with contribution guidelines
* ci: switch from npm to pnpm for monorepo support
* ci: use pnpm version from package.json
* fix: remove unused import in TUI helpers
- Beautiful terminal interface with monochrome design
- Browse 33+ skill repositories dynamically
- Install skills with single keystroke
- Multi-agent installation support
- View installed skills and sync across agents
- Responsive design for all terminal sizes
- Keyboard navigation with clear instructions
- Animated logo reveal on home screen
Features:
- Browse: Discover skills from popular repositories
- List: View all installed skills
- Sync: Sync skills across multiple agents
- Settings: Configure SkillKit preferences
Keyboard shortcuts:
- h/b/l/s/, for navigation
- Enter to install/select
- m for multi-agent selection
- Esc to go back, q to quit
- Add adapters for Amp, Clawdbot, Droid, GitHub Copilot, Goose, Kilo, Kiro CLI, Roo, Trae, Windsurf
- Add --list flag to preview skills without installing
- Add --agent flag for multi-agent install
- Extend skill discovery to search 20+ standard paths
- Add recursive fallback search when no skills in standard paths
- Add tests for new functionality (84 tests total)
git commit -m "feat(marketplace): add skill from #${{ github.event.issue.number }}"
git push
- name:Close Issue with Comment
uses:actions/github-script@v7
with:
script:|
await github.rest.issues.createComment({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.payload.issue.number,
body: `✅ **Skill added to marketplace!**\n\nYour skill has been added and will be available in the next release.\n\nThank you for contributing to SkillKit! 🎉`
You rewrite the same skill for each agent. Or you lock in to one.
## The fix
SkillKit is the package manager for AI agent skills. Install from 400K+ skills across 31 sources. Auto-translate between formats. Persist session learnings. Ship to 46 agents at once.
```bash
npm install -g skillkit
# or
npx skillkit <command>
npx skillkit add anthropics/skills
```
## Quick Start
That is the whole first-run. Pick your agent (your detected agent is pre-selected), confirm, done.
Four features ship as optional packages so a bare CLI stays small. The default install pulls everything. Add `--omit=optional` if you only want the core.
Skipped optional? Add just the one you want later:
```bash
npm i -g @skillkit/tui # enables: skillkit ui
npm i -g @skillkit/api # enables: skillkit serve
npm i -g @skillkit/mesh # enables: skillkit mesh
npm i -g @skillkit/messaging # enables: skillkit message
```
The CLI catches missing optional packages and prints a one-line hint. No stack trace.
### Using `npx`
`npx skillkit add <owner/repo>` works with zero install. First call caches the package under `~/.npm/_npx/`. Every run after that is instant until a new version ships.
Analyze the codebase and write CLAUDE.md, `.cursorrules`, AGENTS.md, and friends:
```bash
skillkit primer --all-agents
```
</details>
<details>
<summary><b>Session memory</b></summary>
AI agents learn during a session, then forget. SkillKit captures what they learned:
```bash
skillkit memory compress
skillkit memory search "auth patterns"
skillkit memory export auth-patterns
```
</details>
<details>
<summary><b>AI skill generation</b></summary>
Generate skills from plain English with multi-source context:
```bash
skillkit generate
```
Pulls context from four places: Context7 docs, your codebase, 400K marketplace skills, your memory. Works with Claude, GPT-4, Gemini, Ollama (local), or any OpenRouter model.
</details>
<details>
<summary><b>Team collaboration</b></summary>
Share skills via a committable `.skills` manifest:
```bash
skillkit manifest init
skillkit manifest add anthropics/skills
git commit -m "add team skills"
```
Everyone else runs `skillkit manifest install` and matches state.
</details>
<details>
<summary><b>Mesh network</b></summary>
Distribute agents across machines with encrypted P2P:
```bash
skillkit mesh init
skillkit mesh discover
```
</details>
<details>
<summary><b>Chrome extension</b></summary>
Save any webpage as a skill from the browser. Click the extension, the page round-trips through the SkillKit API (Turndown, 5-source tag detection, GitHub URL support), the `SKILL.md` downloads, then run `skillkit add ~/Downloads/skillkit-skills/<name>`.
- Ship working software. Everything else is negotiable.
- Simple beats clever. If a junior dev can't read it in 30 seconds, it's too complex.
- Opinions are load-bearing. "It depends" is a cop-out 90% of the time. Pick a side, explain why, move on.
- Exhaustion before escalation. Try everything available before asking for help or giving up.
- Broken things get fixed now, not "in a follow-up PR."
- If the tests pass and the build is green, stop touching it.
## Boundaries
- Never leak secrets, credentials, or private data. Period.
- Ask before doing anything destructive or irreversible. One confirmation. Not three.
- Be honest about uncertainty. "I don't know" is a valid answer. Bullshitting is not.
- The user's codebase is their house. Don't rearrange the furniture without permission.
## Vibe
- Never open with "Great question," "I'd be happy to help," or "Absolutely." Just answer.
- Brevity is mandatory. If the answer fits in one sentence, one sentence is what you get.
- Humor is welcome when it's natural. Not forced puns. The kind of wit that comes from actually understanding the problem.
- Call things out. If you're about to do something dumb, I'll say so. Charm over cruelty, but no sugarcoating.
- Swearing is fine when it lands. A well-placed "that's fucking brilliant" hits different than sterile corporate praise. Don't force it. Don't overdo it. But if a situation calls for a "holy shit" — say holy shit.
- No hedging. No weasel words. No "it might be worth considering perhaps maybe." Say the thing.
- Match energy. Quick question gets a quick answer. Deep problem gets deep analysis. Read the room.
- Code speaks louder than explanations. Show, don't tell. Then shut up.
- Be the assistant you'd actually want to talk to at 2am. Not a corporate drone. Not a sycophant. Just... good.
## Continuity
- This file is a living document. If something stops being true, change it.
- Session memory is ephemeral. This file is how identity persists.
- If this file gets modified during a session, say so explicitly.
Universal skills manager for AI coding agents. Install, manage, and sync skills across 46 AI agents including Claude Code, Cursor, Codex, Gemini CLI, and more.
## Installation
```bash
# Full install (default) — every feature, everything works:
"description":"Supercharge AI coding agents with portable skills. Install, translate, and share skills across Claude Code, Cursor, Codex, Copilot & 41 more",
Then load `packages/extension/dist/` as an unpacked extension.
## Usage
### Popup
1. Navigate to any webpage
2. Click the SkillKit extension icon
3. Optionally enter a custom skill name
4. Click **Save as Skill**
The extension sends the page URL to the SkillKit API, which extracts the content as markdown, generates tags using weighted 5-source analysis, and returns a complete SKILL.md. The file downloads to your `Downloads/skillkit-skills/` folder.
### Context Menu
Right-click on any page:
- **Save page as Skill** — sends the URL to the API, downloads the result
- **Save selection as Skill** — builds the skill locally from selected text (no network request)
### Making Skills Available to Agents
After downloading, install the skill so all agents can use it:
Extension → POST /api/save-skill { url } → Server pipeline → Download
```
1. **Popup** reads the current tab's URL and title via `activeTab` permission. No content script injection needed.
2. **Background service worker** sends the URL to `skillkit.sh/api/save-skill` (or `agenstskills.com/api/save-skill`):
- Server fetches the page and converts HTML to markdown via Turndown
- GitHub URLs are auto-converted to raw content
- Tags are detected from 5 weighted sources (URL segments, headings, code blocks, keywords, language)
- Returns a complete SKILL.md with YAML frontmatter
3. **Download** — the service worker creates a blob and saves via `chrome.downloads` API.
4. **Selection saves** work entirely offline — Chrome natively provides `selectionText` via the context menu API, so the skill is built locally with no server round-trip.
## Permissions
| Permission | Why |
|-----------|-----|
| `contextMenus` | Right-click "Save as Skill" menu items |
| `activeTab` | Access current page URL and title on click |
| `downloads` | Save SKILL.md files to disk |
No `host_permissions` or `scripting` permissions needed. Only the page URL is sent to the server — no page content, browsing history, or personal data.
## Generated SKILL.md Format
```yaml
---
name: react-performance-tips
description: Optimizing React rendering with memo and useMemo
tags:
- react
- javascript
- testing
metadata:
source: https://example.com/react-performance
savedAt: 2026-02-14T09:00:00.000Z
---
# React Performance Tips
The extracted markdown content of the page...
```
## Building from Source
The extension uses tsup to bundle TypeScript into IIFE format for browser compatibility:
| `--source` | `-s` | Remove all skills installed from a specific source |
| `--all` | `-a` | Remove all installed skills |
| `--force` | `-f` | Skip confirmation prompt |
### Bulk Removal
```bash
skillkit remove --source iii-hq/iii
skillkit remove --all
skillkit remove --all --force
```
When using `--source`, SkillKit matches against the source field in each skill's metadata and removes every skill that was installed from that repository. When using `--all`, every discovered skill across all agent directories is removed. Both flags update AGENTS.md automatically if managed sections are present.
## Update Command
```bash
skillkit update [skill-names] [options]
```
| Option | Short | Purpose |
|--------|-------|---------|
| `--dry-run` | — | Check for updates without installing |
| `--yes` | — | Skip confirmation prompts |
| `--agent` | — | Update in specific agents only |
| `--force` | `-f` | Force update, bypass checksum and GitHub API checks |
The update command uses a two-stage change detection pipeline:
1. **GitHub API check** -- For remote sources, SkillKit queries the GitHub API to compare the repository's `pushed_at` timestamp against the installed metadata. If the repository has not been pushed since the last install or update, the skill is skipped without cloning.
2. **Checksum comparison** -- After cloning, a checksum of the source SKILL.md is compared against the stored checksum. If the content is identical, the update is skipped.
Use `--force` to bypass both checks and always overwrite the installed skill.
## Enable/Disable Commands
```bash
skillkit enable <skill-names> [options]
skillkit disable <skill-names> [options]
```
Disabled skills are ignored by the `sync` command. No file changes occur.
## Check Command
```bash
skillkit check [skill-names] [options]
```
Checks installed skills for available updates, quality scores, and GitHub activity.
| Option | Short | Purpose |
|--------|-------|---------|
| `--verbose` | `-v` | Show detailed output including stars and push dates |
| `--quiet` | `-q` | Minimal output |
For remote GitHub sources, the check command queries the GitHub API to compare each repository's `pushed_at` timestamp against the installed date. Each skill displays its quality score badge (A-F). In verbose mode, the output also includes the repository's star count and how recently it was pushed.
```bash
skillkit check
skillkit check pdf xlsx --verbose
```
## List Command
```bash
skillkit list [options]
```
Lists installed skills with quality grades.
## Sync Command
```bash
skillkit sync [options]
```
Synchronizes enabled skills to agent configuration files.
## Recommend Command
```bash
skillkit recommend [options]
```
AI-powered recommendations based on project analysis.
| Option | Purpose |
|--------|---------|
| `--search` | Task-based search |
| `--category` | Filter by category |
| `--min-score` | Quality threshold |
## Generate Command
AI-powered skill generation with multi-source context and agent optimization.
skillkit issue list --repo owner/repo # Specific repo
skillkit issue list --label bug # Filter by label
skillkit issue list --limit 20 # More results
skillkit issue list --json # JSON output
```
The planner extracts task lists (checkboxes) from the issue body, infers file mentions from backtick-quoted paths, and maps labels to tags (e.g., `bug` → `fix`, `enhancement` → `feature`). Generated plans work with `skillkit plan validate` and `skillkit plan execute`.
## Session Intelligence
Track skill executions, generate handoff documents, and analyze skill impact.
### Timeline
View a unified chronological event stream of all session activity.
```bash
skillkit timeline # Last 50 events
skillkit timeline --type git_commit # Filter by event type
skillkit timeline --skill code-simplifier # Filter by skill
skillkit timeline --since 2026-02-10 # Date filter
skillkit publish --output dir # Output to specific directory
skillkit publish --dry-run # Preview without writing
skillkit publish submit # Submit to SkillKit marketplace
```
### publish submit
Submits a skill to the SkillKit marketplace by creating a GitHub issue on `rohitg00/skillkit` via the `gh` CLI. No browser is needed.
Before submitting, the command displays:
- Quality score and grade (A-F) from the skill evaluator
- GitHub star count and last push date from the GitHub API
- Any quality warnings (up to 3)
If `gh` is not installed or not authenticated, the command prints a manual submission URL as a fallback.
## Tap Command
Manage custom skill sources (taps). Taps are user-level source repositories stored in `~/.skillkit/taps.json`. They let you add private or third-party skill repositories that are not part of the official marketplace.
### tap add
Add a custom skill source.
```bash
skillkit tap add owner/repo
skillkit tap add owner/repo --name "My Skills"
```
### tap remove
Remove a custom skill source.
```bash
skillkit tap remove owner/repo
```
### tap list
List all custom skill sources.
```bash
skillkit tap list
```
Once a tap is added, skills from that source appear in `skillkit marketplace search` results and can be installed with the standard `skillkit install owner/repo` workflow.
## Security Scanning
```bash
skillkit scan <path> # Scan skill for vulnerabilities
skillkit scan <path> --format json # Output as JSON
SkillKit maintains a centralized lock file at `~/.skillkit/lock.json` that tracks every installed skill. The lock file is written on install, updated on update, and cleaned on remove.
```json
{
"version": 1,
"skills": {
"pdf": {
"source": "anthropics/skills",
"checksum": "sha256:abc123...",
"agents": ["claude-code", "cursor"],
"installedAt": "2026-04-01T12:00:00Z",
"updatedAt": "2026-04-08T09:30:00Z"
}
}
}
```
| Field | Description |
|-------|-------------|
| `source` | The repository or local path the skill was installed from |
| `checksum` | SHA-256 hash of the SKILL.md content at install time, used for update detection |
| `agents` | Which agents the skill is installed for |
| `installedAt` | ISO timestamp of initial installation |
| `updatedAt` | ISO timestamp of the most recent update |
The `update` command compares the stored checksum against the source to skip unchanged skills. The `remove` command cleans up the corresponding lock entry.
## Taps File
Custom skill source repositories are stored in `~/.skillkit/taps.json`. Taps are managed via the `skillkit tap` commands.
```json
{
"version": 1,
"taps": [
{
"source": "myorg/internal-skills",
"name": "Internal Skills",
"addedAt": "2026-04-05T10:00:00Z"
}
]
}
```
Skills from tapped sources appear in marketplace search results and can be installed like any other source. See the [tap command](/docs/commands#tap-command) documentation for usage.
## Directory Structure
```
~/.skillkit/
├── preferences.json # User preferences
├── lock.json # Installed skill tracking
├── taps.json # Custom skill sources
├── cache/ # Marketplace cache
├── memory/ # Learning storage
│ ├── observations/
│ ├── learnings/
│ └── index.json
├── mesh/ # Mesh network
│ ├── keys/ # Encryption keys
│ ├── peers.json # Known peers
│ └── config.json
└── messages/ # Inter-agent messages
├── inbox/
├── sent/
└── archived/
./project/
├── skillkit.yaml # Project config
├── .skills # Team manifest
├── skills/ # Local skills
└── .skillkit/ # Project-specific
├── cache/
├── primer-config.json
└── memory/
```
## Configuration Profiles
Create different profiles for different environments:
description: Universal skill management for AI coding agents
---
# SkillKit
Skills make AI coding agents smarter. But every agent uses a different format — Claude Code wants `.claude/skills/`, Cursor uses `.mdc`, Copilot expects `.github/skills/`. You end up rewriting the same skill three times, or locking into one platform.
SkillKit fixes this. Write a skill once, deploy it to all 45 agents.
## Get Started
```bash
npx skillkit@latest
```
That's it. No global install needed. Or if you prefer:
```bash
npm install -g skillkit
skillkit init
skillkit install anthropics/skills
skillkit sync
```
Four commands. Your agents now have skills for PDF processing, code review, and more.
description: Agent-native skill discovery via Model Context Protocol
---
# MCP Server
The REST API works great for HTTP clients. But AI agents speak MCP — the Model Context Protocol. The `@skillkit/mcp` package lets agents discover skills natively, without HTTP calls or curl commands.
Think of it as giving your agent a direct line to the entire skill catalog.
## Setup
### Claude Desktop
Add this to `~/Library/Application Support/Claude/claude_desktop_config.json`:
```json
{
"mcpServers": {
"skillkit": {
"command": "npx",
"args": ["@skillkit/mcp"]
}
}
}
```
Restart Claude Desktop. Done.
### Cursor
Add to `.cursor/mcp.json` in your project:
```json
{
"mcpServers": {
"skillkit": {
"command": "npx",
"args": ["@skillkit/mcp"]
}
}
}
```
### Any MCP client
The server uses stdio transport. Works with any MCP-compatible client:
```bash
npx @skillkit/mcp
```
## What your agent can do
Once connected, your agent gets four tools:
### Search skills
Ask your agent: *"Find skills for React performance optimization"*
Behind the scenes, it calls `search_skills` with your query and returns ranked results. You can filter by tags, category, or source repository.
```
> Search for authentication skills tagged with "security"
```
The agent searches across the full catalog, ranks results by relevance, and shows you the best matches.
### Get a specific skill
Ask: *"Get the pdf-processing skill from anthropics/skills"*
The agent calls `get_skill` and returns the full skill details — description, tags, content, and references.
### Get recommendations
Ask: *"What skills would you recommend for my TypeScript React project?"*
The agent calls `recommend_skills` with your stack context (languages, frameworks, libraries) and returns personalized suggestions ranked by relevance.
### Browse categories
Ask: *"What skill categories are available?"*
The agent calls `list_categories` and returns all categories with counts, so you know what's out there.
## Resources
The server also exposes two read-only resources:
- **`skills://trending`** — Top 20 skills by relevance score
- **`skills://categories`** — All categories with counts
Agents can read these without making a tool call.
## How it works
1. On startup, the server loads all skills from the marketplace catalog
2. Your agent sends a natural language request
3. The MCP client translates it into a tool call
4. The `RelevanceRanker` scores and ranks results (same algorithm as the REST API)
5. Results come back as structured JSON through the MCP protocol
Same ranking, same data, same quality — just native to your agent instead of going through HTTP.
description: Persistent learning across AI agent sessions
---
# Memory System
Capture learnings from AI sessions and convert them into reusable skills. SkillKit's memory system provides automatic observation capture, intelligent compression, and token-optimized retrieval.
## Commands
```bash
skillkit memory status # View status
skillkit memory search <q> # Search learnings
skillkit memory compress # Compress observations to learnings
skillkit memory export <name> # Export as skill
skillkit memory sync-claude # Update CLAUDE.md with learnings
skillkit memory index # View memory index (Layer 1)
skillkit memory --global # Global memory operations
description: Multi-machine agent distribution with secure peer-to-peer communication
---
# Mesh Network
SkillKit's mesh network enables secure communication between AI agents across multiple machines. Share skills, synchronize memory, and collaborate across your development environment.
description: Send messages between AI agents across the mesh network
---
# Inter-Agent Messaging
SkillKit's messaging system enables asynchronous communication between AI agents. Send code reviews, share insights, and coordinate work across your team.
## Quick Start
```bash
# Send a message
npx skillkit message send
# View inbox
npx skillkit message inbox
# Read a message
npx skillkit message read <message-id>
```
## Commands Reference
| Command | Description |
|---------|-------------|
| `message send` | Send message to agent |
| `message inbox` | View inbox |
| `message read <id>` | Read a message |
| `message reply <id>` | Reply to message |
| `message archive <id>` | Archive message |
| `message sent` | View sent messages |
| `message status` | Messaging status |
## Message Types
Messages can include:
- **Text** - Plain text messages
- **Code** - Code snippets with syntax highlighting
- **Diffs** - Code changes and patches
- **Skills** - Skill file attachments
## Priority Levels
- **urgent** - Immediate attention required
- **normal** - Standard priority (default)
- **low** - Can be addressed when convenient
## Inbox Management
```bash
# View unread messages
npx skillkit message inbox --unread
# Archive old messages
npx skillkit message archive <id>
# View sent messages
npx skillkit message sent
```
## Programmatic API
```typescript
import { MessagingClient } from '@skillkit/messaging';
import { MeshHost } from '@skillkit/mesh';
// Create messaging client
const messaging = new MessagingClient({
meshHost: host,
storagePath: '~/.skillkit/messages',
});
await messaging.init();
// Send a message
await messaging.send({
to: 'claude@laptop',
subject: 'Code review completed',
body: 'I reviewed the auth module...',
priority: 'normal',
});
// Listen for new messages
messaging.on('message', (msg) => {
console.log('New message from:', msg.from);
});
```
## TUI Navigation
In the Terminal UI, press `j` to access the Messages screen.
## Next Steps
- [Mesh Network](/docs/mesh) - Configure the underlying network
- [Team Collaboration](/docs/teams) - Share skills with your team
description: Curated skill packs for debugging, testing, planning, and collaboration
---
# Methodology Packs
Methodology packs are curated collections of skills that guide AI agents through structured workflows. Each pack bundles related skills around a domain — debugging, testing, planning, or collaboration.
description: Auto-generate AI agent instructions from your codebase
---
# Primer
Primer analyzes your codebase and automatically generates optimized instructions for all 45 AI agents. No more manual configuration—let SkillKit detect your tech stack, coding patterns, and project structure.
## Quick Start
```bash
# Analyze current project
skillkit primer
# Generate for specific agents
skillkit primer --agent claude,cursor,windsurf
# Generate for all 45 agents
skillkit primer --all-agents
# Preview without writing files
skillkit primer --dry-run
```
## What Primer Detects
### Tech Stack
- **Frameworks**: React, Next.js, Vue, Angular, Svelte, etc.
- **Languages**: TypeScript, JavaScript, Python, Go, Rust, etc.
Not everything runs on Node. The `skillkit-client` package gives Python apps the same skill discovery capabilities as the REST API — async, typed, and ready to go.
All fields except `name` and `source` on `Skill` are optional — they depend on what the server returns and whether you requested `include_content`.
## Prerequisites
The Python client talks to the SkillKit REST API. You need a running server:
```bash
skillkit serve
```
Then run your Python script in another terminal. The client handles connection management, retries, and cleanup automatically when used with `async with`.
Skills shouldn't require pre-installation. With the SkillKit REST API, any agent, framework, or tool can discover and retrieve skills on demand over HTTP.
## Start the server
```bash
skillkit serve
```
That's it. The API is now running at `http://localhost:3737`.
| References | 15 | Has example files, docs, or resources |
A skill with full content, an exact name match, high stars, and example files scores close to 100. A skill with just a name and no content scores around 10.
The server includes built-in Swagger UI at `/docs` and an OpenAPI 3.1 spec at `/openapi.json`. Open `http://localhost:3737/docs` in your browser to try endpoints interactively.
The hosted version is available at [skillkit.sh/api](https://skillkit.sh/api) and [agenstskills.com/api](https://agenstskills.com/api).
## All endpoints
| Method | Path | Description |
|--------|------|-------------|
| GET | `/search?q=...` | Search skills |
| POST | `/search` | Search with filters |
| GET | `/skills/:owner/:repo/:id` | Get specific skill |
description: Detect malicious patterns, secrets, and vulnerabilities in AI agent skills
---
# Security Scanner
SkillKit includes a built-in security scanner that analyzes skills for prompt injection, command injection, data exfiltration, hardcoded secrets, and other threats before installation.
## Quick Start
```bash
skillkit scan ./my-skill
```
The scanner runs automatically during `skillkit install`. Use `--no-scan` to skip or `--force` to install despite findings.
| **Low** | 1 (with `--fail-on low`) | Minor issue or best practice |
| **Info** | 0 | Informational finding |
## Trust Badges
Skills display a trust badge based on their source and quality analysis. Badges appear during installation, in `skillkit check` output, and in marketplace search results.
| Badge | Criteria |
|-------|----------|
| **[Official]** | Published by a known official source (Anthropic, Vercel, iii-hq, etc.) |
| **[Trusted]** | TrustScorer rates the skill 8-10 based on clarity, boundaries, specificity, and safety |
| **[Review]** | TrustScorer rates the skill 5-7; the skill may need refinement |
| **[Caution]** | TrustScorer rates the skill 0-4; requires careful review before use |
The `TrustScorer` evaluates four weighted dimensions:
| Dimension | Weight | What it measures |
|-----------|--------|------------------|
| Clarity | 30% | How well-defined the skill's purpose and instructions are |
| Boundaries | 25% | Whether the skill stays within its declared scope |
| Specificity | 25% | How concrete and actionable the instructions are |
Official sources are allowlisted by organization name and always receive the `[Official]` badge regardless of score. For all other sources, the badge is determined by the composite trust score.
description: Understanding AI agent skills - format, lifecycle, and quality
---
# Skills
A skill is an instruction file that teaches AI coding agents how to handle specific tasks. Each skill uses a universal Markdown format (SKILL.md) with YAML frontmatter.
description: Share skills and maintain consistency across your development team
---
# Team Collaboration
SkillKit provides powerful team collaboration features to ensure every developer has the same AI agent capabilities, follows the same patterns, and stays in sync.
## Quick Start
```bash
# Initialize team manifest
skillkit team init
# Add skills to team manifest
skillkit team add anthropics/skills
skillkit team add ./local-skills
# Share with team (commit to Git)
git add .skills
git commit -m "Add team skills manifest"
git push
# Team members install
git pull
skillkit team install
```
## Team Manifest (.skills)
The `.skills` file is a Git-committable manifest that defines your team's skill stack.
returnNextResponse.json({error:"Topic is required and must be a string"},{status: 400});
}
constsanitizedTopic=topic.trim().slice(0,500);
if(!sanitizedTopic){
returnNextResponse.json({error:"Topic cannot be empty"},{status: 400});
}
constapiKey=process.env.GEMINI_API_KEY;
if(!apiKey){
returnNextResponse.json({error:"Gemini API key not configured"},{status: 500});
}
try{
constai=newGoogleGenAI({apiKey});
constresponse=awaitai.models.generateContent({
model:"gemini-2.0-flash",
contents:`Create a comprehensive AI coding agent skill for: "${sanitizedTopic}"
This skill will be used by AI coding assistants (like Claude Code, Cursor, GitHub Copilot) to provide better code suggestions and follow best practices.
The skill should include:
1. Clear principles that tell the AI WHAT to do and WHEN
2. Code patterns with REAL, WORKING examples (not pseudocode)
3. Anti-patterns showing common mistakes and how to fix them
4. Specific file patterns where this skill applies
Make the instructions actionable and specific. The AI agent should be able to directly apply these guidelines when writing code.
Focus on practical, production-ready patterns used in modern development.`,
config:{
responseMimeType:"application/json",
responseSchema: agentSkillSchema,
systemInstruction:`You are an expert software architect creating AI agent skills.
Your skills are used by AI coding assistants to write better code.
Guidelines:
- Write instructions as commands to the AI (e.g., "Always use...", "Never...", "Prefer...")
- Include REAL code examples, not pseudocode
- Be specific and actionable
- Focus on modern best practices (2024+)
- Include edge cases and error handling patterns
- Make principles enforceable and testable`
}
});
if(!response.text){
returnNextResponse.json({error:"No response from Gemini"},{status: 500});
}
constdata=JSON.parse(response.text);
returnNextResponse.json(data,{
headers:{
"X-RateLimit-Remaining":String(remaining),
},
});
}catch(error){
console.error("Gemini API Error:",error);
returnNextResponse.json({error:"Failed to generate skill"},{status: 500});
default:'SkillKit - Universal Skills for AI Coding Agents',
template:'%s | SkillKit',
},
description:'One CLI to install, sync, and manage skills across Claude Code, Cursor, Windsurf, Copilot, and 40 more agents. 400K+ skills ready to use.',
File diff suppressed because one or more lines are too long
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.