agentleFS
nerymurillohndnerymurillohnd/the-claude-factory/.claude/references/skill.md

references

Skill1 starsChanged yesterdayMITView on GitHub

How to create a plugin skill's SKILL.md from .claude/templates/skill-template.md. plugins/<plugin>/skills/<skill>/SKILL.md, plus any supporting files beside it (references/, scripts/, assets/). Read 2026-11-10, Claude Code docs skills: This repository:

12 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When a plugin gains a skill, and when a skill's purpose or triggers change.

    line 11
  2. Very specific:

    Read 2026-11-10, Claude Code docs skills:

    line 15
  3. Very specific:

    Create plugins/<plugin>/skills/<skill>/ and copy the template to SKILL.md.

    line 34
  4. Distinctive:

    Add only the optional fields the skill needs.

    line 36
  5. Distinctive:

    Write the body; move long material to files beside it and link them.

    line 37
  6. Distinctive:

    Never take another plugin's skill as a model.

    line 53
  7. Distinctive:

    Never add a frontmatter field the skill does not use.

    line 63
  8. Distinctive:

    Never put a secret or a machine path in a skill.

    line 64

“Common” is measured: kept by at least 6 in 10 of the 9 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

Skill

How to create a plugin skill's SKILL.md from .claude/templates/skill-template.md.

Produces

plugins/<plugin>/skills/<skill>/SKILL.md, plus any supporting files beside it (references/, scripts/, assets/).

When

  • When a plugin gains a skill, and when a skill's purpose or triggers change.

Sources

Read 2026-11-10, Claude Code docs skills:

  • Frontmatter is read only when the opening --- is the file's first line; YAML that doesn't parse loads the skill with no fields set.
  • name is optional and defaults to the directory name; description is recommended: "Put the key use case first: the combined description and when_to_use text is truncated at 1,536 characters in the skill listing."
  • Optional fields: when_to_use, argument-hint, arguments, disable-model-invocation, user-invocable, allowed-tools, disallowed-tools, model, effort, context, agent, background, hooks, paths, shell, metadata, license, compatibility.
  • Outside Claude Code (claude.ai uploads, the Skills API) only name, description, license, compatibility, metadata and allowed-tools are allowed; any other field fails the upload with a hard error.
  • "Keep SKILL.md under 500 lines. Move detailed reference material to separate files."

Placeholders

ValueMeaningTrue value from
{{skillName}}Kebab-case name, identical to the directoryThe skill's directory
{{skillDescription}}What the skill does and when to use it, key use case firstThe skill's own files
{{skillTitle}}The H1 headingThe skill's purpose
{{skillInstructions}}The body Claude followsThe skill's procedure

Procedure

  1. Create plugins/<plugin>/skills/<skill>/ and copy the template to SKILL.md; the --- stays on line 1.
  2. Fill name and description; delete the instruction comment.
  3. Add only the optional fields the skill needs. For a plugin that reaches Claude, keep to the portable subset, or the skill fails to upload there.
  4. Write the body; move long material to files beside it and link them.
  5. Run the checks.

Rules

Official:

  • Frontmatter on line 1, valid YAML.
  • description plus when_to_use within 1,536 characters, key use case first.
  • SKILL.md under 500 lines.

This repository:

  • The description is written from the skill's own files, imperative, with no quoted trigger phrases.
  • A colon goes only after a field name. No value holds a colon or is wrapped in quotes: reword the sentence instead, so every value stays plain text that strict YAML parsers read.
  • name equals the directory name.
  • Never take another plugin's skill as a model.

Checks

  • claude plugin validate --strict plugins/<plugin>.
  • claude plugin details plugins/<plugin>: the skill is listed, with its size.
  • python3 scripts/validators/skill-names-consistency.py and python3 scripts/validators/skill-size.py.

Never

  • Never add a frontmatter field the skill does not use.
  • Never put a secret or a machine path in a skill.

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.