Skip to content

Agent skills

Agent skills are folders of instructions and scripts that teach Claude Code, Codex, Cursor and other agents a task. Each one lists what it can run and how to install it.

Showing 1,201–1,230 of 3,419

  1. Turn the current response into a clear, immediately executable next action.

    • Open source
    • 120 GitHub stars
  2. Reconnect a drifting discussion to the original intent.

    • Open source
    • 120 GitHub stars
  3. Identify the assumptions the current thinking relies on.

    • Open source
    • 120 GitHub stars
  4. Identify meaningful directions for exploring the current thinking.

    • Open source
    • 120 GitHub stars
  5. Find strong counterarguments, counterexamples, or failure cases for the current thinking.

    • Open source
    • 120 GitHub stars
  6. Create a resumable checkpoint of our current thinking.

    • Open source
    • 120 GitHub stars
  7. Ask the single most important clarification question before answering.

    • Open source
    • 120 GitHub stars
  8. Compare competing ideas or interpretations in the current thinking.

    • Open source
    • 120 GitHub stars
  9. Use this skill to generate well-branded interfaces and assets for Honeycomb (shared AI agent memory, a Legion Code × Activeloop product), ei

    • Open source
    • 118 GitHub stars
    • Design and media
  10. Use before starting non-trivial work that Honeycomb may already have prior context for (a past decision, a stated convention, where somethin

    • Open source
    • 118 GitHub stars
    • Knowledge and memory
  11. Honeycomb persistent shared memory for the OpenClaw harness — auto-capture and recall across sessions via the local Honeycomb daemon.

    • Open source
    • 118 GitHub stars
  12. Z.AI CLI providing: - Vision: image/video analysis, OCR, UI-to-code, error diagnosis (GLM-4.6V) - Search: real-time web search with domain/r

    • Open source
    • 111 GitHub stars
    • Files and documents
  13. Explore code structure, architecture, files, directories, and component relationships using the OpenTrace knowledge graph

    • Open source
    • 111 GitHub stars
  14. Run the full OpenTrace knowledge-graph pipeline over a folder or repo — index, clustering, highlight surfacing, and the markdown / graphml /

    • Open source
    • 111 GitHub stars
    • Files and documents
  15. Build, package, and verify DesktopManager NuGet, PowerShell module, and CLI/MCP artefacts through the repo-standard PowerForge entrypoints

    • Open source
    • 108 GitHub stars
  16. Operate and validate Windows desktop state through DesktopManager using MCP first and CLI second

    • Open source
    • 108 GitHub stars
  17. Inspect Windows desktop applications and verify requested UI workflows with DesktopManager MCP, screenshots, and window or control evidence

    • Open source
    • 108 GitHub stars
  18. Use whenever a task needs an API key, token, password, or other secret — before asking the user for a value, before writing any .env or hard

    • Open source
    • 87 GitHub stars
  19. Audit PHP engine deserialization surface (unserialize, session decoders, WDDX, phar metadata, custom ce->unserialize handlers, Serializable

    • Open source
    • 85 GitHub stars
  20. Use when the user wants to check, understand, or harden the security of their Mac, or asks whether their laptop is secure and what to fix

    • Open source
    • 82 GitHub stars
  21. Solve coding, debugging, research, analysis, planning, and other agentic work with the smallest verified solution, adaptive effort, minimal

    • Open source
    • 73 GitHub stars
  22. Process academic PDFs into structured Obsidian literature notes and 9-node critical-thinking canvases

    • Open source
    • 66 GitHub stars
    • Knowledge and memory
  23. This skill should be used when the user asks about 'wt', 'worktree', 'worktrees', 'wt create', 'wt checkout', 'wt list', 'wt remove', 'wt pr

    • Open source
    • 64 GitHub stars
  24. Write a production-ready spec for an AI agent: scope, tools and permissions, autonomy levels, stop rules, escalation, failure modes, evals,

    • Open source
    • 59 GitHub stars
  25. Turn a shaped solution into a breadboard: user actions, UI affordances, system responsibilities, and wiring so it can be sliced and implemen

    • Open source
    • 59 GitHub stars
  26. Use when asked to "run a design sprint", "5-day sprint", "prototype in a week", "test ideas before building", or "Jake Knapp sprint"

    • Open source
    • 59 GitHub stars
    • Design and media
  27. Use when asked to "growth loops", "build a growth engine", "design a viral loop", "create a content loop", "move beyond paid acquisition", o

    • Open source
    • 59 GitHub stars