by diegosouzapw
Public repository of AI coding skills, curated improved best-practice skills, and runtime surfaces for CLI, API, MCP, and A2A.
# Add to your Claude Code skills
git clone https://github.com/diegosouzapw/awesome-omni-skillsGuides for using ai agents skills like awesome-omni-skills.
Last scanned: 7/31/2026
{
"issues": [
{
"type": "npm-audit",
"message": "@hono/node-server: Node.js Adapter for Hono: Path traversal in `serve-static` on Windows via encoded backslash (`%5C`)",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "@modelcontextprotocol/sdk: Vulnerability found",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "body-parser: body-parser vulnerable to denial of service when invalid limit value silently disables size enforcement",
"severity": "low"
},
{
"type": "npm-audit",
"message": "fast-uri: fast-uri vulnerable to host confusion via literal backslash authority delimiter",
"severity": "high"
},
{
"type": "npm-audit",
"message": "hono: Hono: IP Restriction bypasses static deny rules for non-canonical IPv6 ",
"severity": "high"
},
{
"type": "npm-audit",
"message": "postcss: PostCSS: Arbitrary file read and information disclosure via attacker-controlled sourceMappingURL in CSS comments",
"severity": "high"
},
{
"type": "npm-audit",
"message": "qs: qs has a remotely triggerable DoS: qs.stringify crashes with TypeError on null/undefined entries in comma-format arrays when encodeValuesOnly is set",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "tmp: tmp has Path Traversal via unsanitized prefix/postfix that enables directory escape",
"severity": "high"
},
{
"type": "npm-audit",
"message": "vite: launch-editor: NTLMv2 hash disclosure via UNC path handling on Windows",
"severity": "high"
},
{
"type": "npm-audit",
"message": "ws: ws: Uninitialized memory disclosure",
"severity": "high"
}
],
"status": "WARNING",
"scannedAt": "2026-07-31T06:43:28.698Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}awesome-omni-skills is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by diegosouzapw. Public repository of AI coding skills, curated improved best-practice skills, and runtime surfaces for CLI, API, MCP, and A2A. It has 100 GitHub stars.
awesome-omni-skills returned warnings in SkillsLLM's automated security scan. It has no critical vulnerabilities, but review the flagged issues in the Security Report section before adding it to your workflow.
Clone the repository with "git clone https://github.com/diegosouzapw/awesome-omni-skills" and add it to your Claude Code skills directory (see the Installation section above).
awesome-omni-skills is primarily written in Python. It is open-source under diegosouzapw on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh awesome-omni-skills against similar tools.
No comments yet. Be the first to share your thoughts!
Requires a passing catalog security scan. Resolve the flagged issues and resubmit to enable featuring.
The skill catalog that installs itself.
CLI · API · MCP · A2A — all from a single npx command.
🚀 4715 skills · 7 bundles · 9 install clients · 16 MCP clients · 4 runtime surfaces
⚡ Install in 1 min · 🛠️ Choose your tool · 📖 CLI Guide · 📦 Bundles · 🔌 Runtime · 💡 Why Awesome Omni Skills
| Metric | Value | |
|---|---|---|
| 📦 | Published Skills | 4715 native skills across 17 active categories |
| 🎯 | Bundles | 7 fully backed curated bundles |
| 🖥️ | Install Clients | 9 install-capable AI coding assistants |
| 🔌 | MCP Clients | 16 MCP config-capable clients across 33 first-class targets |
| 🔐 | Curated Output | 1623 enhanced English derivatives in skills_omni |
| 🧪 | Validation Mix | 4073 passed · 642 warn · 0 errors |
| 📋 | Current Release | v0.12.9 |
🔍 Searched for AI coding skills, Claude Code skills, Cursor skills, Codex CLI skills, Gemini CLI skills, Antigravity skills, or installable
SKILL.mdlibraries? You're in the right place. 👇
Awesome Omni Skills is an installable skill catalog and runtime for AI coding assistants. At its core, it is a public repository of reusable SKILL.md playbooks, a curated repository of improved best-practice derivatives, and the CLI, API, MCP, and A2A runtime surfaces that distribute and operate the same catalog.
Native intake enters skills/ in two ways: direct contributor PRs and reviewed repository-based external-import PRs. Curated English-only derivatives are then proposed separately into skills_omni/ by the private enhancement pipeline.
| Component | Description |
|---|---|
| 🧠 Skills | Reusable SKILL.md-based playbooks for AI assistants, including native intake from direct contributor PRs and reviewed repository-based external-import PRs |
| 📦 Manifests | Generated JSON manifests, bundles, and archives |
| 🧭 Guided Install | Interactive TTY and visual terminal install flows |
| 🌐 Catalog API | Read-only HTTP API for search, discovery, and downloads |
| 🔌 MCP Server | Discovery, recommendation, and client-aware config tooling |
| 🤖 A2A Runtime | Agent-to-agent task orchestration |
| ✨ Enhancement Pipeline | Private enhancer publishes curated English derivatives into skills_omni/ |
# Open the visual shell in an interactive terminal
npx awesome-omni-skills
# Or force the text guided installer explicitly
npx awesome-omni-skills install --guided
test -d ~/.gemini/antigravity/skills && echo "✅ Skills installed"
💬 "Use
@brainstormingto plan a SaaS MVP."💬 "Use
@api-designto review this endpoint design."💬 "Use
@debuggingto isolate this regression."
| 🎯 Goal | Bundle | Command |
|---|---|---|
| General engineering | essentials |
npx awesome-omni-skills --bundle essentials |
| Product + app delivery | full-stack |
npx awesome-omni-skills --bundle full-stack |
| Design systems | design |
npx awesome-omni-skills --bundle design |
| Security review | security |
npx awesome-omni-skills --bundle security |
| Infra and release | devops |
npx awesome-omni-skills --bundle devops |
| LLM applications | ai-engineer |
npx awesome-omni-skills --bundle ai-engineer |
| OSS maintenance | oss-maintainer |
npx awesome-omni-skills --bundle oss-maintainer |
Before comparing bundles or picking an install path, understanding these five building blocks helps:
| Concept | What it means |
|---|---|
| 🧠 Skills | Reusable SKILL.md playbooks that teach an assistant how to execute a workflow well |
| 📦 Catalog Artifacts | Generated JSON and archive outputs enabling search, comparison, download, and install |
| 🔌 MCP Config | Client-side configuration for assistants to discover Awesome Omni Skills through MCP tools |
| 🤖 A2A Runtime | Agent-to-agent orchestration for discovery, recommendation, and install-plan handoff |
| ✨ Curated Output | skills_omni/ — the Awesome Omni Skills-maintained enhanced surface, separate from native upstream intake |
📝 Native/Curated policy:
skills/accepts native upstream intake in any language, whether submitted directly or proposed by the private external-sync runtimeskills_omni/is always curated and published in Englishskills_omni/is a one-way surface and does not loop back into native intake- direct contributor PRs and repository-based external intake share the same validator, metadata refresh, and enhancer path
Not just "another repository with skills in folders." Awesome Omni Skills has a stronger contract and a broader runtime surface.
| If you want… | 📁 Typical skills repo | ✨ Awesome Omni Skills |
|---|---|---|
| Install into a real assistant | Manual copy or custom script | npx awesome-omni-skills, guided install, visual UI, selective --skill and --bundle |
| Search and compare skills | Browse markdown manually | Generated catalog, filtering, bundle planning, search, compare, and recommendation |
| Use the same data across tools | Separate logic per tool | Shared manifests and catalog for CLI, API, MCP, and A2A |
| Configure MCP clients | Hand-edit files | config-mcp, local sidecar previews, generated recipes, and allowlisted writes |
| Trust repository content | Minimal or ad hoc linting | Critical static security gate blocks unsafe patterns such as remote content piped into shell or instructions that try to expose prompts, secrets, or hidden context |
| Trust releases | Best-effort packaging | Checksums, signed archives, ClamAV + VirusTotal verified release CI, and publish preflight |
| Curate community intake | Whatever lands stays as-is | Native intake in skills/, curated English derivatives in skills_omni/ with attribution |
These skills follow the SKILL.md model and can be used as a normal repository, but the package also installs and configures them across a broad surface:
9 install-capable clients · 16 MCP config-capable clients
| Tool | Type | Invocation Example | Install Path |
|---|---|---|---|
| 🟢 Claude Code | CLI | Use brainstorming to plan a feature |
~/.claude/skills |
| 🔵 Cursor | IDE | @brainstorming help me plan a feature |
~/.cursor/skills |
| 🟡 Gemini CLI | CLI |