504 curated AI agent skills from 19 verified sources. One install, every major coding assistant.
npx @flitzrrr/agent-skills installDetects installed platforms and provisions skills for each. Target a single platform with:
npx @flitzrrr/agent-skills install <platform>Where <platform> is one of: vscode, codex, antigravity, opencode.
Manual setup (git clone)
git clone --recurse-submodules git@github.com:flitzrrr/agent-skills.git
cd agent-skills
# Symlink into target platform(s):
ln -sf $(pwd)/skills/* ~/.copilot/skills/ # VS Code / GitHub Copilot
ln -sf $(pwd)/skills/* ~/.codex/skills/ # Codex
ln -sf $(pwd)/skills/* ~/.gemini/antigravity/skills/ # Antigravity
ln -sf $(pwd)/skills/* ~/.config/opencode/skills/ # OpenCode
# Pull upstream updates:
git submodule update --remote --mergeAll CLI commands
| Command | Description |
|---|---|
npx @flitzrrr/agent-skills install |
Install for all detected platforms |
npx @flitzrrr/agent-skills install <platform> |
Install for a single platform |
npx @flitzrrr/agent-skills update |
Pull latest upstream skills |
npx @flitzrrr/agent-skills list |
List all available skills |
| Platform | Discovery Mechanism | Install Method |
|---|---|---|
| VS Code (GitHub Copilot) | .github/copilot-instructions.md + ~/.copilot/skills/ |
install vscode |
| Claude Code | CLAUDE.md (project-level) |
Clone into project |
| Codex | AGENTS.md |
install codex |
| Cursor | .cursorrules (project-level) |
Clone into workspace |
| Lovable | .lovable |
Clone into workspace |
| Windsurf | AGENTS.md |
Clone into workspace |
| Antigravity | ~/.gemini/antigravity/skills/ |
install antigravity |
| OpenCode | AGENTS.md + skills/ |
install opencode |
| Any agent | Read skills/<name>/SKILL.md directly |
Manual |
| Category | Examples | Sources |
|---|---|---|
| Security and code review | security-review, code-review, find-bugs, gha-security-review |
Sentry, Trail of Bits |
| Git workflow | commit, create-branch, pr-writer, iterate-pr |
Sentry |
| Project planning | create-plan, resume-plan, execute-work-package, generate-handover |
DasDigitaleMomentum |
| Document generation | anthropic-pdf, anthropic-docx, anthropic-pptx, anthropic-xlsx |
Anthropic |
| Frontend and React | vercel-react-best-practices, vercel-web-design-guidelines, anthropic-frontend-design |
Vercel, Anthropic |
| Infrastructure | aws-lambda, terraform-code-generation, cloudflare-wrangler |
AWS, HashiCorp, Cloudflare |
| Finance | finance-equity-research, finance-ib-advisory, finance-wealth-mgmt |
Anthropic |
| Science and ML | scientific-bioinformatics, scientific-alphafold, scientific-chembl |
K-Dense-AI |
| Marketing and SEO | content-strategy, seo-audit, product-description-seo |
Coreyhaines, Bhanunamikaze |
| Developer tools | anthropic-mcp-builder, anthropic-webapp-testing, systematic-debugging |
Anthropic, Sentry |
| Django | django-access-review, django-perf-review |
Sentry |
| Deployment | vercel-deploy-to-vercel, stitch-shadcn-ui |
Vercel, Google |
See CHEATSHEET.md for a decision guide on which skill to use for a given task.
| Source | Skills | Focus |
|---|---|---|
| K-Dense-AI/claude-scientific-skills | 175 | Science, bioinformatics, ML, chemistry |
| trailofbits/skills | 60 | Security auditing, static analysis, smart contracts |
| anthropics/financial-services-plugins | 56 | Finance: IB, equity research, PE, wealth mgmt |
| MoizIbnYousaf/Ai-Agent-Skills | 48 | Database design, debugging, code patterns |
| coreyhaines31/marketingskills | 33 | Marketing: SEO, email, content, analytics |
| getsentry/skills | 24 | Security, code review, Git workflow, Django |
| itsmostafa/aws-agent-skills | 18 | AWS infrastructure and services |
| anthropics/skills | 17 | Document gen, creative design, MCP |
| hashicorp/agent-skills | 14 | Terraform generation |
| expo/skills | 11 | Expo design and deployment |
| cloudflare/skills | 9 | Workers, Durable Objects, MCP |
| DasDigitaleMomentum/opencode-processing-skills | 9 | Multi-session planning |
| google-labs-code/stitch-skills | 7 | Design-to-code, shadcn/ui, Remotion |
| vercel-labs/agent-skills | 5 | React, Next.js, web design |
| callstackincubator/agent-skills | 5 | React Native, GitHub workflows |
| JackyST0/awesome-agent-skills | 5 | General-purpose agent patterns |
| stripe/ai | 2 | Stripe best practices |
| Bhanunamikaze/Agentic-SEO-Skill | 1 | SEO auditing (16 sub-skills) |
| supabase/agent-skills | 1 | Postgres best practices |
New sources are discovered weekly via GitHub Actions and security-scanned before integration.
agent-skills/
skills/ 504 entries (symlinks to vendor + local forks)
vendor/ 20 Git submodules (upstream sources)
bin/ CLI and build scripts
docs/ GitHub Pages catalog and project documentation
.github/ CI workflows and copilot-instructions.md
Skills are exposed as a flat directory under skills/. Most entries are symlinks into vendor/ submodules. Skills with local extensions (e.g., execute-work-package with multi-transport support) are maintained as real directories alongside the symlinks.
Skills are namespaced by source to avoid collisions: anthropic-pdf, tob-static-analysis, cloudflare-wrangler, terraform-code-generation, scientific-bioinformatics, aws-lambda, finance-equity-research, callstack-react-native-best-practices. Sentry and marketing skills are un-prefixed (code-review, content-strategy).
| Workflow | Trigger | Purpose |
|---|---|---|
| Skill Discovery | Weekly | Find trending skill repos, security-scan, auto-add |
| Submodule Update | Weekly | Pull upstream changes, open PR |
| MegaLinter | Push / PR | Lint markdown, YAML, JSON |
| npm Publish | Tag v* |
Publish to npm, create GitHub Release |
The execute-work-package skill supports three transport modes for delegating implementation to a sub-agent:
| Transport | Mechanism | When to use |
|---|---|---|
| MCP via l4l-oci (default) | l4l-oci exposes 10 MCP tools for the gated blueprint-gate-execute lifecycle | Any IDE with MCP support |
| Fresh Agent | IDE spawns a new sub-agent per step (no external server) | Fallback when l4l-oci is not configured |
| Stateful Session | Sub-agent session resumed across steps | OpenCode only |
When l4l-oci MCP tools are available (create_handle, generate_blueprint, submit_gate, execute_handle, get_digest), the skill uses them automatically. The bundled scripts/start-l4l-oci.sh auto-starts the server if needed. See the l4l-oci setup guide for configuration.
sequenceDiagram
participant User
participant Primary as Primary Agent<br/>(IDE)
participant MCP as l4l-oci MCP Server
participant Sub as Sub-LLM<br/>(configurable)
User->>Primary: Task with scope and DoD
rect rgb(240, 245, 250)
Note over Primary,Sub: Blueprint Phase
Primary->>MCP: create_handle(project_root)
MCP-->>Primary: handle_id
Primary->>MCP: generate_blueprint(handle_id, prompt)
MCP->>Sub: Generate Execution Blueprint
Sub-->>MCP: Blueprint (ordered steps)
Primary->>MCP: get_blueprint(handle_id)
MCP-->>Primary: Blueprint for review
end
Primary->>User: Present Blueprint for review
User->>Primary: Approve
rect rgb(240, 250, 240)
Note over Primary,Sub: Gate
Primary->>MCP: submit_gate(handle_id, "accept")
MCP-->>Primary: Gated
end
rect rgb(250, 245, 240)
Note over Primary,Sub: Execute Phase
Primary->>MCP: execute_handle(handle_id)
MCP->>Sub: Implement Blueprint steps
Sub->>Sub: Edit files, run verification
Sub-->>MCP: Execution Digest
Primary->>MCP: get_digest(handle_id)
MCP-->>Primary: Digest (outcome, files, verify result)
end
Primary->>User: Report results
MIT -- applies to the CLI, workflows, and documentation in this repository.
Each vendored submodule in vendor/ retains its own original license.