Skip to content

Instruction-only agent skills

2,929 instruction-only agent skills, out of 3,419 agent skills listed. These skills contain instructions and reference files but no scripts, so they run no code of their own. Each one lists the tools it asks for and how to install it in Claude Code, Codex, Cursor and other agents.

Showing 2,911–2,929 of 2,929

  1. Use when an agent's performance is uneven and you want to understand where it excels or struggles

    • Open source
    • 14 GitHub stars
  2. Use when a term is ambiguous or project-specific and needs a canonical definition

    • Open source
    • 14 GitHub stars
  3. Use after major skill edits or when definitions feel inconsistent

    • Open source
    • 14 GitHub stars
  4. Use when you have a plan document and want autonomous implementation until it's done

    • Open source
    • 14 GitHub stars
  5. Use when reviewing a skill/agent/rule and you need to understand why it looks the way it does

    • Open source
    • 14 GitHub stars
  6. Use after a human jam session to turn agreements into code

    • Open source
    • 14 GitHub stars
  7. Use when you need deep understanding of a focused area — not broad exploration

    • Open source
    • 14 GitHub stars
  8. Capture the orchestrator's mental model for a structured session transition

    • Open source
    • 14 GitHub stars
  9. Use after parallel bounded-context implementation to verify contracts still hold

    • Open source
    • 14 GitHub stars
  10. Use when you need multiple perspectives on a topic — brainstorming, requirements clarification, or pressure-testing an idea before committin

    • Open source
    • 14 GitHub stars
  11. Use when something works correctly but is too slow, too memory-heavy, or too resource-intensive

    • Open source
    • 14 GitHub stars
  12. Use before building high-stakes features — anything touching money, auth, data integrity, or security

    • Open source
    • 14 GitHub stars
  13. Use after subagent implementation, before merging PRs, or when reviewing commit ranges

    • Open source
    • 14 GitHub stars
  14. Interactive code review as a panel discussion: an implementer agent defends decisions while a reviewer surfaces findings, with specialist re

    • Open source
    • 14 GitHub stars
  15. Run a session health diagnostic to assess context load, scope drift, and quality degradation

    • Open source
    • 14 GitHub stars
    • Health and science
  16. Use when you need a complete design before coding — multiple valid approaches, architectural alignment needed, or complex features

    • Open source
    • 14 GitHub stars
  17. Use at session start or when you need to orient

    • Open source
    • 14 GitHub stars
  18. Use before a multi-human Event Storming session to prepare independent findings

    • Open source
    • 14 GitHub stars
  19. Use when a feature crosses system boundaries and integration risk is high

    • Open source
    • 14 GitHub stars