awesome-agent-skills
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 11 days ago
- Low visibility — Only 5 GitHub stars
Code Pass
- Code scan — Scanned 4 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
No AI report is available for this listing yet.
Skills for AI coding agents: code review, debugging, security audits, refactoring, cleaning up AI-written code and text, and more.
Awesome Agent Skills
Skills for AI coding agents: code review, test writing, design docs, debugging, security audits, refactoring, and cleaning up AI-written code and text. Each skill is a folder with a SKILL.md in the Agent Skills format — install by copying it into your agent's skills directory.
Compatible with Claude Code, Claude.ai, OpenAI Codex, Gemini CLI, Cursor, GitHub Copilot, opencode, Amp, and any other agent that supports the standard. See Install for exact paths.
Contents
Quick start
Clone once, then symlink every skill into your agents — a later git pull keeps them all in sync:
git clone https://github.com/khasky/awesome-agent-skills.git
cd awesome-agent-skills
npx skills add ./skills -g -s "*" -a claude-code codex gemini-cli -y
The skills CLI symlinks all 28 skills globally into Claude Code, Codex, and Gemini CLI, pointing them at your clone. Change -a to pick agents (-a "*" installs to every detected agent); add --copy to write independent copies instead of symlinks.
Keep them current:
cd awesome-agent-skills && git pull # updates every symlinked skill in place
Re-run the skills add command to pick up skills added since. Then just ask the agent naturally ("review this diff against main") or invoke a skill explicitly (/awesome-code-review in Claude Code and Cursor, $awesome-code-review in Codex).
Prefer not to clone? npx skills add khasky/awesome-agent-skills installs straight from GitHub — but it won't auto-sync with git pull.
Skills
Code review
| Skill | What it does |
|---|---|
| awesome-code-review | Reviews diffs and PRs for correctness, security, and team standards |
| awesome-code-review-feedback | Responds to review feedback: verify before implementing, clarify, push back with reasoning |
Code quality and refactoring
| Skill | What it does |
|---|---|
| awesome-code-standards | Naming, structure, and patterns for consistent code across a team |
| awesome-code-cleanup | Repo-wide cleanup of AI-like code noise: comments by default, plus a read-only audit mode and a refactor mode for vague names, dead code, and over-abstraction — behavior-preserving |
| awesome-dependency-upgrade | Executes dependency upgrades safely: risk-classified batches, changelog-driven majors, overrides with removal conditions, verification between steps |
Testing
| Skill | What it does |
|---|---|
| awesome-test-writing | Designs and writes tests that catch real regressions: placement ladder, behavior-first assertions, characterization tests, property/fuzz for parsers — every test proven able to fail |
Design and planning
| Skill | What it does |
|---|---|
| awesome-design-doc | Produces design docs and ADRs: requirements and numbers first, real alternatives, a recommendation tied to requirements, non-goals and rollout |
| awesome-api-design | Designs or reviews HTTP API shape before code: resource modeling, versioning by layering, cursor pagination, idempotency keys |
Debugging and reliability
| Skill | What it does |
|---|---|
| awesome-bug-fix | Reproduce → isolate root cause → fix → verify; no random edits |
| awesome-root-cause | Root-cause analysis for incidents and process problems with no failing test: 5-Whys, fishbone, PDCA, one-page A3 |
| awesome-error-standards | Consistent error handling, retries, and user-facing messages |
| awesome-logging-standards | Structured logging, levels, and PII handling |
Audits
| Skill | What it does |
|---|---|
| awesome-architecture-audit | Read-only whole-project audit: architecture and module boundaries, docs-vs-code fidelity, YAGNI/KISS/SOLID, and extensibility (build-from-source, add-a-provider), with a SHIP/FIX/BLOCK verdict and a prioritized recommendations report |
| awesome-security-audit | Checks for injection, secrets, auth issues, dependency CVEs, CI/CD pipeline exposure, and crypto misuse |
| awesome-pentest | Authorization-gated penetration-test engagement: scoping and rules of engagement, recon, attack-surface mapping, testing, exploitation-to-proof, and a findings report with CWE/CVSS and retest — drives the audit siblings and follows PTES/OWASP WSTG/NIST 800-115 |
| awesome-leak-audit | Keeps a public client (extension, app, SPA, CLI) from leaking backend internals; client hardening |
| awesome-accessibility-audit | WCAG-oriented a11y checks and fixes |
| awesome-seo-audit | Read-only SEO + AI-discoverability audit: technical SEO, programmatic-page safety, agent/LLM readability, with a SHIP/FIX/BLOCK verdict |
| awesome-performance-audit | Read-only server/runtime performance and reliability audit: event loop, streams/backpressure, memory diagnostics, graceful shutdown, resilience topology (circuit breakers, retry budgets, queue bounds), with a SHIP/FIX/BLOCK verdict |
| awesome-db-audit | Read-only database-layer audit: schema anti-patterns (EAV, imprecise types), query/index fit, integrity and concurrency, migration hygiene and tenancy, with a SHIP/FIX/BLOCK verdict |
| awesome-dependency-audit | Read-only supply-chain audit of the dependency graph: lockfile discipline, typosquats and hallucinated package names, install-script exposure, CVE reachability, licenses — with a SHIP/FIX/BLOCK verdict |
| awesome-landing-audit | Read-only structural audit of landing/marketing pages: single CTA, form friction, message match, trust elements, CLS-safe banners — mechanics, not copy |
Git and repository operations
| Skill | What it does |
|---|---|
| awesome-git-history-reset | Wipes a repo's git history to a single Initial commit and force-pushes — safely: access checks, verified mirror backup, secret scan, and a confirmation gate before anything irreversible |
Writing and text
| Skill | What it does |
|---|---|
| awesome-humanize-en | Removes signs of AI generation from English text: 36 patterns, regex markers, source-fabrication checks |
| awesome-document-style | Line-edits Markdown into clear, specific, publication-ready prose |
| awesome-grammar-check | Advisory copy-edit — grammar, logic, and flow issues as suggestions, without rewriting the text |
| awesome-translate-ru-en | Russian → English translation that preserves structure, formatting, and the author's voice |
| awesome-style-mimic | Learns a website's writing style into a reusable style guide (live-browser deep crawl), then rewrites documents in that voice with a cross-document consistency pass |
Install
The Quick start covers Claude Code, Codex, and Gemini CLI. For any other agent — or to install by hand without the CLI — a skill is just a folder: copy it into the directory your agent reads. Most agents also read the shared .agents/skills/ path, so one copy can serve several tools at once. To stay in sync with git pull, symlink from your clone instead of copying (ln -s).
| Agent | Project path | Global path | Docs |
|---|---|---|---|
| Claude Code | .claude/skills/ |
~/.claude/skills/ |
docs |
| OpenAI Codex | .agents/skills/ |
~/.agents/skills/ |
docs |
| Gemini CLI | .gemini/skills/ or .agents/skills/ |
~/.gemini/skills/ or ~/.agents/skills/ |
docs |
| Cursor | .cursor/skills/ or .agents/skills/ |
~/.cursor/skills/ or ~/.agents/skills/ |
docs |
| GitHub Copilot | .github/skills/ or .agents/skills/ |
~/.copilot/skills/ or ~/.agents/skills/ |
docs |
| opencode | .opencode/skills/ or .agents/skills/ |
~/.config/opencode/skills/ |
docs |
| Amp | .agents/skills/ |
~/.agents/skills/ |
docs |
| Windsurf | .windsurf/skills/ |
~/.codeium/windsurf/skills/ |
docs |
| Antigravity | .agents/skills/ (legacy .agent/skills/) |
~/.gemini/antigravity/skills/ |
docs |
- Claude.ai (web): zip a skill folder and upload it under Settings → Skills.
- Gemini CLI can also install straight from a repo URL:
gemini skills install <repo-url> --consent. - Restart or reload the agent after copying so it picks up new skills.
Usage examples
Skills trigger on intent — plain requests work. Explicit mentions make the choice deterministic:
- "Use awesome-code-review on the diff against
main: Critical / Suggestions / Nice to have, with file:line references." - "Here are the review comments — use awesome-code-review-feedback: agree, clarify, or push back with evidence."
- "awesome-bug-fix for this error: reproduce first, root cause before any fix — [paste logs]."
- "awesome-security-audit the files touched by this change (auth, injection, secrets)."
- "Audit this repo with awesome-leak-audit before we open-source it."
- "awesome-humanize-en: remove the AI voice from this draft."
- "awesome-translate-ru-en: mirror
docs-ru/intodocs-en/— same structure, translate each file."
They also chain: implement → awesome-code-cleanup in audit mode (read-only findings) → apply the cleanup → awesome-code-review before merge.
Skill format
Each skill follows the Agent Skills specification:
skills/<skill-name>/
├── SKILL.md # required: YAML frontmatter (name, description) + instructions
├── references/ # optional: detailed docs the agent loads on demand
└── scripts/ # optional: helper scripts
No README.md inside a skill folder. A skill folder holds only what the agent reads: SKILL.md and the files it names. What each skill is for and when to reach for it belongs here in the root README (the tables above), so a reader compares skills in one place instead of opening 28 folders. Where a skill ships references/ or scripts/, SKILL.md itself maps them — an agent that skips the map skips the files.
The frontmatter description tells the agent when to activate the skill; the body loads only after activation, and references/ files only when needed — so a large skill still costs little context until used.
Two rating vocabularies are shared across the collection, so two reports never mean different things by the same word:
- Findings —
Critical / High / Medium / Low / Informational, rated on impact and reachability. A skill may truncate the scale from the bottom — top four whereInformationalcarries no meaning (architecture, database), top three whereLowdoesn't either (accessibility, copy-editing, landing mechanics) — and its output section states which tiers it uses. Never reorder or rename tiers. - Verdict —
SHIP / FIX / BLOCK, for read-only audits that gate a release (landing, performance, SEO), always paired withNOT ASSESSEDfor anything that could not be checked.
awesome-code-review keeps its own reviewer-comment buckets (Critical / Suggestions / Nice to have) because those are addressed to an author, not to a release gate.
Design principles
- Self-contained — every skill is complete in this repo; no chasing external links or docs.
- Few, not many — a small set that covers repeated engineering work, not a hundred micro-skills nobody remembers.
- Portable — plain
SKILL.mdper the open spec, stack-agnostic, nothing vendor-specific baked in. - Verification-first — skills end with the check that proves the claim: run the command, read the output, then say "done".
Which layer to install
A rule is a standing constraint the agent honors without being asked; a skill is a procedure you invoke, with phases and an output contract. The two layers overlap by design: rules/code-review.md in Awesome AGENTS.md sets the bar every review must meet, awesome-code-review here runs the review and produces the report. Install both — rules keep everyday work in line, skills handle the jobs you name.
Related
Part of a set of agent tooling — pick the layer you need:
- Awesome Agent Skills — this repo: portable
SKILL.mdskills every agent loads — code review, debugging, security and leak audits, code and text cleanup. - Awesome AGENTS.md — the base, tool-agnostic ruleset every agent imports (one
AGENTS.md). - Agent MCP Integrations — MCP servers that connect agents to browsers, cloud, databases, infra, and domain APIs.
- Claude Code Token Optimization — the token-efficiency layer (RTK, LSP, Context7,
codebase-memory-mcp, claude-mem, Caveman, Ponytail). - Claude Code Security Audit — the layered security-audit workflow (deep audit, continuous guardrails, scanners).
Contributing
A new skill is justified only when no existing one covers the trigger — otherwise it becomes a section in the skill that already fires. CONTRIBUTING.md has the frontmatter contract, the two body templates, the shared rating vocabularies, and the checks CI runs.
License
Released under the MIT license. The awesome-humanize-en skill adapts MIT-licensed material from humanizer-ru by Vladimir-Human — with credit kept inline where it's used.
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found