agentleFS
Sign inStart free
danielrosehilldanielrosehill/Claude-Israel-Agent-Skills-Plugin/skills/update-plugin-readme/SKILL.md

update-plugin-readme

Skill2 starsChanged 24 days agoMITView on GitHub

Use when the user wants to refresh the README of this plugin (or any Claude Code skills plugin) so that its "Skills" section reflects the current set of skills under `skills/*/SKILL.md`. Reads each skill's frontmatter (`name`, `description`, optional `category`), groups by category, sorts alphabetically within each group, and rewrites the README's Skills section between the `<!-- SKILLS:START -->` and `<!-- SKILLS:END -->` markers. Preserves everything else in the README (header, installation, license). A helper script at `scripts/generate-skills-section.py` does the extraction and emits the block on stdout; the skill is responsible for splicing it into the README. Trigger phrases - "update the plugin README", "regenerate the skills table", "refresh skills list on the plugin README", "the plugin README is out of date", "add the new skill to the README".

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run python3 scripts/generate-skills-section.py <plugin-repo-path> (default: the current plugin repo).

    line 19
  2. Distinctive:

    Remove any "Planned scope" section that contains items now implemented.

    line 22
  3. Very specific:

    If a SKILL.md lacks a category field, the script infers from folder name:

    line 42
  4. Very specific:

    Prefer adding the category field to the skill's frontmatter over relying on inference.

    line 52

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Update Plugin README

Regenerates the "Skills" section of a Claude Code skills plugin README from the actual contents of its skills/ directory.

When to run

  • A new skill has been added to the plugin's skills/ folder.
  • A skill's frontmatter description has been edited.
  • A skill has been removed.
  • The README drifts from reality (e.g. lists planned skills that already ship).

Procedure

  1. Run python3 scripts/generate-skills-section.py <plugin-repo-path> (default: the current plugin repo).
  2. The script emits a block starting with <!-- SKILLS:START --> and ending with <!-- SKILLS:END -->. Capture its stdout.
  3. Open README.md. Replace everything between those two markers with the new block. If the markers don't exist, locate the existing ## Skills heading and replace that section (up to the next ## heading) with the new block wrapped in markers; if there's no ## Skills heading, insert the block after the README's opening paragraph.
  4. Remove any "Planned scope" section that contains items now implemented. Leave genuinely unimplemented items alone but move them into an "Ideas" section if still plausible; drop them otherwise.
  5. Show the user a diff before committing.

Category taxonomy

The helper script uses folder-name inference. Preferred explicit category values (add to a skill's SKILL.md frontmatter if inference gets it wrong):

CategoryScope
Government & CivicPostal service, municipal reports, vehicle licensing, national services
HealthcareHMOs, medicines, medical appointments
Emergency PreparednessShelters, Home Front Command guidelines, alerts
LocalizationHebrew language, fonts, translation
Media & InformationNews feeds, media lookups
Meta / ToolingSkills that manage the plugin itself

Category ordering in the README is fixed (Government & Civic → Healthcare → Emergency Preparedness → Localization → Media & Information → Meta / Tooling → Other).

Fallback category inference rules

If a SKILL.md lacks a category field, the script infers from folder name:

  • *-municipality-*, israel-post-*, *-government-* → Government & Civic
  • maccabi-*, clalit-*, meuhedet-*, leumit-*, *-medicine-*, *-medical-* → Healthcare
  • *-shelter-*, miklat*, home-front-*, pikud-*, *-alert-* → Emergency Preparedness
  • hebrew-*, *-translation, *-fonts-* → Localization
  • *-news-*, *-rss-* → Media & Information
  • add-skill-*, update-plugin-*, install-*-plugin, *-plugin-* → Meta / Tooling
  • Otherwise → Other (prompt the user)

Prefer adding the category field to the skill's frontmatter over relying on inference.

Description sanitisation

The full description frontmatter for a skill is discovery-oriented (contains "Use when..." preamble and trigger phrases). The README version is a plain, user-facing sentence. The helper script strips the Use when the user wants to prefix, truncates at the first sentence boundary, cuts everything from Trigger phrases onward, capitalises the first letter, and drops the trailing period.

Out of scope

  • Regenerating the rest of the README (header, installation, license, related plugins). Those are hand-maintained.
  • Editing individual SKILL.md files. This skill only reads them.

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