VoltAgent

VoltAgent/awesome-agent-skills

MIT AI & ML

A curated collection of 1000+ agent skills from official dev teams and the community, compatible with Claude Code, Codex, Gemini CLI, Cursor, and more.

27.7k stars
3k forks
recent
GitHub +587 / week

27.7k

Stars

3k

Forks

57

Open issues

30

Contributors

AI Analysis

Awesome Agent Skills is a curated collection of 1,500+ skills for AI agent platforms (Claude Code, Gemini CLI, Cursor, GitHub Copilot, and others), sourced from official development teams and community contributors rather than AI-generated. It serves developers integrating AI agents into applications who need vetted, production-grade skills across APIs, services, and workflows from partners like Anthropic, Google, Stripe, Cloudflare, and Figma.

AI & ML Developer Tool Discovery value: 4/10
Documentation 8/10
Activity 9/10
Community 9/10
Code quality 5/10

Inferred from signals mentioned in the README (tests, CI, type safety) — not a review of the actual code.

Overall score 8/10

AI's overall editorial judgment — not an average of the bars above, can weigh other factors too.

agent-skills ai-agents claude-code skill-collection multi-platform
Actively maintained Well documented MIT licensed Popular Community favorite Beginner friendly Production ready
Deep Analysis · Based on README and public signals
2w ago

Curated library of 1,400+ agent skills from real engineering teams, compatible with major AI coding tools

awesome-agent-skills is a community-maintained collection of reusable agent skill definitions (likely CLAUDE.md, .cursorrules, or equivalent config files) for AI-assisted development tools. It targets developers using Claude Code, Cursor, Gemini CLI, Codex, and similar agentic coding environments who want pre-built, vetted behavioral instructions contributed by recognized engineering teams at Anthropic, Google, Vercel, Stripe, Cloudflare, and others. Its primary value is saving developers the time of writing effective agent instructions from scratch.

Origin

Created October 2025, during a period of rapid adoption of agentic coding tools. Appears to have grown quickly by aggregating skills from official dev teams, distinguishing itself from bulk AI-generated repositories in the same emerging category.

Growth

726 stars in 7 days as of June 2026 indicates sustained momentum roughly 8 months after creation. Growth is likely driven by: the proliferation of agentic coding tools making this category newly relevant, credibility from official team contributions (Anthropic, Google, Vercel), and community contribution dynamics typical of curated 'awesome' lists. The similar-repo landscape shows a cluster of competing repos with higher star counts, suggesting the category itself is experiencing a broad surge.

In production

The README lists contributions from recognizable engineering organizations (Anthropic, Google, Vercel, Stripe, Cloudflare, Netlify, Hugging Face, Sentry, Figma, etc.), which suggests at least some of these skills are used in real development workflows. However, direct evidence of end-user adoption at scale — such as download metrics, CLI install counts, or user testimonials — is not provided in the available README. Adoption not verified beyond stated contributor affiliations.

Code analysis
Architecture

Appears to be a content-only repository — likely a collection of text files (Markdown, YAML, or similar) organized by contributor and compatible tool, with no compiled code. The 'Language: unknown' metadata supports this. Likely structured as a directory of skill files with a README index linking to an external hosting platform (officialskills.sh).

Tests

not documented in README

Maintenance

Last push June 23, 2026 — one day before evaluation date — indicates very active maintenance. Commit history likely reflects ongoing community contributions and curation. Sponsorship presence and Discord community suggest organized stewardship rather than solo maintenance.

Honest verdict

ADOPT IF: you actively use Claude Code, Cursor, Gemini CLI, or similar agentic tools and want vetted, reusable behavioral instructions without writing them from scratch — especially for frameworks or services with official team contributions. AVOID IF: you need programmatic skill management, versioning, or validation tooling; this is a content collection, not a software library. MONITOR IF: you are evaluating the agent-skills ecosystem broadly — this repo tracks which official teams are publishing skills and can serve as a signal for where the industry is standardizing.

Independent dimensions

Mainstream potential

5/10

Technical importance

4/10

Adoption evidence

3/10

Risks
  • The 'awesome list' format has low defensibility — a competitor repo with more stars (VoltAgent's own openclaw collection has 50k) could consolidate attention quickly.
  • Curation quality is difficult to verify externally; the claim of 'hand-picked, not AI-slop' is asserted but not auditable from metadata alone.
  • The category itself is immature and standards for agent skills (file format, schema, compatibility) are not yet stabilized across tools, meaning skills may become stale or incompatible as tools evolve.
  • Heavy reliance on third-party officialskills.sh domain for skill links creates a single point of failure outside GitHub's control.
  • The fragmented landscape of similar repos (multiple competitors with higher star counts) suggests no clear community consolidation point yet, which may limit long-term canonical status.
Prediction

Likely to remain a useful reference hub as long as agentic coding tools grow, but faces ongoing fragmentation risk from the cluster of competing repos. May consolidate with or be superseded by tool-native skill marketplaces if Claude Code, Cursor, or others build first-party discovery.

0 found this helpful

Newsletter

Get analyses like this every Monday

Free weekly digest of the most interesting open-source discoveries.

Languages

No language breakdown available.

Information

License
MIT
Last updated
1w ago
Created
9mo ago
Analyzed with
anthropic/claude-haiku-4-5

Stars over time

Loading…

Contributors over time

Top 100 contributors only — repos with more will plateau at 100.

Loading…

Recent releases

No releases published yet.

Similar repos

sickn33

sickn33/agentic-awesome-skills

Agentic Awesome Skills is a searchable, installable library of 1,936+ reusable...

42.7k Python Dev Tools
libukai

libukai/awesome-agent-skills

This is a curated guide and resource hub for AI agent skills—a lightweight...

4.8k Python AI & ML
heilcheng

heilcheng/awesome-agent-skills

Awesome Agent Skills is a curated directory of reusable AI agent skill...

6k TypeScript AI & ML
alirezarezvani

alirezarezvani/claude-skills

A comprehensive open-source library of 355+ production-ready skills and plugins...

21.9k Python Dev Tools
VoltAgent

VoltAgent/awesome-claude-code-subagents

A curated collection of 154+ specialized Claude Code subagents organized across...

23.1k Shell AI & ML
vs. alternatives
sickn33/antigravity-awesome-skills

Has 41,521 stars vs 26,236 — significantly more visible in the ecosystem. Possibly tied to the 'Antigravity' tool mentioned as a compatible target. May have broader reach but unclear if it matches the curation quality focus this repo claims.

VoltAgent/awesome-openclaw-skills

Same VoltAgent organization with 50,552 stars — likely a sister repo targeting a specific tool (OpenClaw). Higher star count suggests that specialization may be more discoverable than a multi-tool general collection.

heilcheng/awesome-agent-skills

5,792 stars — a smaller competing collection with similar naming. Likely less organized or newer. This repo appears more comprehensive based on contributor list breadth.

alirezarezvani/claude-skills

18,902 stars, Python-based — likely more narrowly focused on Claude-specific skills. May be more precise for Claude-only users but less useful across multiple AI coding tools.

rohitg00/skillkit

1,242 stars, TypeScript — appears to be a tooling layer rather than a pure content collection. Likely solves a related but different problem (generating or managing skills programmatically vs. curating them).