Agent Skills Hub

504 curated AI agent skills from 19 verified sources. One install, every major coding assistant.

Full catalog | Decision guide

Quick Start

npx @flitzrrr/agent-skills install

Detects 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.

<details> <summary>Manual setup (git clone)</summary>

git clone --recurse-submodules [email protected]: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 --merge

</details>

<details> <summary>All CLI commands</summary>

</details>

Platform Support

Skill Categories

See CHEATSHEET.md for a decision guide on which skill to use for a given task.

Sources

New sources are discovered weekly via GitHub Actions and security-scanned before integration.

Architecture

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.

Naming Convention

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).

Automation

MCP Execution Backend (l4l-oci)

The execute-work-package skill supports three transport modes for delegating implementation to a sub-agent:

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.

Lifecycle

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

License

MIT -- applies to the CLI, workflows, and documentation in this repository. Each vendored submodule in vendor/ retains its own original license.