agentleFS
Sign inStart free
yuusakuriyuusakuri/agent-skills/skills/using-agent-skills/SKILL.md

using-agent-skills

Skill2 starsChanged 21 days agoMITView on GitHub

Helps choose among the installed agent Skills without replacing their native description-based discovery. Use when the user explicitly asks which Skill to use, requests an overview of the catalog, or several Skill descriptions remain equally applicable. Do not invoke merely because a session started.

49 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use constraint-driven-development and doubt-driven-development.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MrCipherSmithSkill

    constraint-driven-development

    Use when a project needs a persistent, measurable quality bar — starting a project without documented standards, 'set up constraints', or when coverage/perf/a11y thresholds get re-debated per-PR. Produces CONSTRAINTS.md with enforceable numbers, ratchets, and a guard against silent weakening. NOT for throwaway prototypes or a one-off CI setup.

    Also in 15 projects that use this

  • seikaikyoSkill

    doubt-driven-development

    對非平凡決策啟動新鮮 context 對抗審查（找碴不背書），在修正還便宜的時候抓出錯誤方向。適用：高風險改動（production、資安敏感邏輯、不可逆操作）、不熟的程式碼、要宣稱「這樣是安全的 / 可行的」之前。機械性操作與一行修改不適用。

    Also in 38 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 44 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the available Skill names and descriptions supplied by the agent.

    line 16
  2. Distinctive:

    Prefer the Skill that produces the exact requested artifact.

    line 18
  3. Distinctive:

    Use multiple Skills only when their outputs are sequential dependencies.

    line 19
  4. Distinctive:

    Never invent or invoke a Skill name that is not installed.

    line 20
  5. Distinctive:

    When a more specific installed Skill matches the requested named artifact or framework, prefer it over a broad workflow Skill.

    line 37
  6. Distinctive:

    When the user asks which Skill applies, return:

    line 53

“Common” is measured: kept by at least 6 in 10 of the 44 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Using Agent Skills

Choose the smallest set of installed Skills that owns the requested outcome.

Use native discovery first

Codex and Claude Code discover each Skill from the name and description in that Skill's SKILL.md. This Skill is optional guidance, not a required root manifest and not a dispatcher that must run before other Skills.

  1. Read the available Skill names and descriptions supplied by the agent.
  2. Match the user's requested outcome and current lifecycle stage.
  3. Prefer the Skill that produces the exact requested artifact.
  4. Use multiple Skills only when their outputs are sequential dependencies.
  5. Never invent or invoke a Skill name that is not installed.

Resolve common overlaps

NeedPrimary SkillBoundary
Clarify an ambiguous requestinterview-meUse brainstorming after the problem is understood and options are needed.
Explore solution optionsbrainstormingUse spec-driven-development once the implementation behavior is ready to define.
Define implementation behaviorspec-driven-developmentUse planning-and-task-breakdown only after the specification is stable.
Break implementation into stepsplanning-and-task-breakdownIt does not own product requirements or formal system documents.
Create a product requirements documentdeliver-prdUse document-architecture for formal requirements, specifications, designs, and proposals.
Create a formal system documentdocument-architectureIt routes to one document type and keeps Markdown canonical unless Word is requested.
Record an architecture decisiondocumentation-and-adrsAn ADR records why a decision was made; a technical specification records the selected design.
Implement a changeincremental-implementationCombine with test-driven-development when behavior changes.
Review before mergecode-review-and-qualityAdd security or performance review only when those risks are present.
Prepare a releaseshipping-and-launchUse release-notes only for the user-facing change summary.

When a more specific installed Skill matches the requested named artifact or framework, prefer it over a broad workflow Skill.

Sequence dependent Skills

A typical implementation can use:

spec-driven-development → planning-and-task-breakdown → incremental-implementation + test-driven-development → code-review-and-quality → shipping-and-launch

This is an example, not a mandatory pipeline. Skip any stage whose owned artifact or verification is already complete.

Report the choice

When the user asks which Skill applies, return:

  • the selected Skill or ordered set;
  • the outcome each one owns;
  • why nearby Skills were not selected;
  • any required input that is still missing.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free