agentleFS
majdghithanmajdghithan/agent-skills/skills/skill-authoring/SKILL.md

skill-authoring

Skill7 starsChanged 4 months agoMITView on GitHub

Author, structure, verify, and publish a high-quality agent skill (SKILL.md). Use whenever the user wants to "write a skill", "create a skill for X", turn expertise or a workflow into a reusable skill, improve an existing SKILL.md, or publish a skill to their repo. Covers the SKILL.md format, writing a strong description, progressive disclosure with reference files, the detect-context-first pattern, fact-verification discipline, testing, and the publish workflow.

  • Commits and pushes

89 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use cloud-api-recipe-authoring, codebuddy-ide-mcp-upgrade and 10 more.

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.
  • TencentCloudBaseSkill

    cloud-api-recipe-authoring

    Author or revise a cloud-api-operations recipe (config/source/skills/cloud-api-operations/references/recipes/). Encodes why the long tail of control-plane capabilities is deliberately NOT wrapped as MCP/CLI tools, the API-first admission test for whether a capability deserves a recipe, the negative and positive lists, the fixed recipe structure, the writing rules distilled from writing-for-agents, the user-facing copy red lines, and the product regeneration/verification pipeline. Also routes contributors (and their agents) to the contribution flow and the pre-PR self-test protocol in references/contributing.md. Use when adding a recipe, judging whether a scenario belongs in the recipe set, updating the recipes index, re-checking existing recipes against the admission test, or onboarding someone who wants to submit one.

    Also in 3 projects that use this

  • TencentCloudBaseSkill

    codebuddy-ide-mcp-upgrade

    升级 CodeBuddy IDE(genie 扩展)内置的 CloudBase MCP,以及 MCP 发版时同步 IDE 侧白名单。当用户提到「更新 IDE 里的 MCP」「内置 MCP 版本太老」「IDE 集成的 CloudBase 功能不足」「改工具白名单 toolWhiteList」「把新 bundle 打进 CodeBuddy」「白名单漂移」「MCP 发版要同步什么」时使用。覆盖:解包定位内置 bundle 与内嵌配置、重新构建 mcp bundle、生成新的工具白名单与系统提示词、安全注入 IDE 并备份、用 MCP 协议验证工具清单、人工端到端验收、一键回滚。

    Also in 3 projects that use this

  • DIL8654Skill

    api-contract-review

    Review ASP.NET Core Web API contracts for HTTP semantics, versioning, backward compatibility, and response consistency. Use when the user asks to review an API, check endpoints, validate REST behavior, or before releasing contract changes.

    Also in 5 projects that use this

  • zaxbysauceSkill

    pr-review-fix

    Codex adapter for addressing pull request review feedback in opencode-swarm. Use when the user asks to fix PR review comments, requested changes, reviewer findings, CI-review notes, or a pasted review summary with low false-positive tolerance.

    Also in 3 projects that use this

  • albertdobmeyerSkill

    git-workflows

    Advanced git operations beyond add/commit/push. Use when rebasing, bisecting bugs, using worktrees for parallel development, recovering with reflog, managing subtrees/submodules, resolving merge conflicts, cherry-picking across branches, or working with monorepos.

    Also in 3 projects that use this

  • rogue-dev-studioSkill

    frontend-engineering

    Expert frontend application engineering: accessible UI structure, state and data-fetching discipline, form/validation UX, responsive layouts, performance budgets, and integration with API contracts. Use when building or refining SPA/web screens, design-system consumption, client-side routing, or fixing UI reliability (loading, empty, and error states).

    Also in 3 projects that use this

  • +6 more from TencentCloudBase

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 79 edited versions we read keep these.

  • 2

    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:

    No marketing, no filler, no restating the obvious.

    line 42
  2. Very specific:

    Create skills/<name>/SKILL.md (+ references/ if multi-file).

    line 55
  3. Distinctive:

    Add a row to the repo README.md skills table.

    line 56
  4. Very specific:

    Commit as Majd Ghithan <majd.ghithan20@gmail.com> (never the session email.

    line 57

“Common” is measured: kept by at least 6 in 10 of the 79 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 authoring - write a great, publishable skill

A skill is a SKILL.md file (YAML frontmatter + markdown instructions) that an agent loads when its work matches. Optionally it bundles supporting files (references/*.md, scripts/) loaded on demand. Good skills are specific, accurate, and lean. This skill is how to write one.

1. Decide it should exist

Write a skill when there's real, reusable know-how an agent would otherwise get wrong or generic: a framework's version-specific rules, a migration's gotchas, a self-hosting workflow, a house style. Don't write one for a one-off task, or for something the model already does well. Check the ecosystem first (npx skills find <topic>) - if a great one exists, install it instead; if only generic ones exist, that's your gap.

2. Frontmatter - name + description (the description is critical)

---
name: kebab-case-name          # matches the folder name
description: <what it does> + <when to use it> + <trigger phrases> + <when NOT to use>
---

The description is how the agent decides to activate the skill - spend real effort on it. Pack it with concrete trigger phrases the user might say, the situations it covers, and an explicit "use when… / do not use for…" boundary. A vague description means the skill never fires (or fires wrongly).

3. Structure - lean SKILL.md, detail in references (progressive disclosure)

  • Keep SKILL.md short and scannable. It's loaded every time the skill activates - don't bloat the context. Put the always-true guidance here; push deep or bulky detail into references/<topic>.md that the agent reads only when needed.
  • When the topic has variants (versions, dialects, platforms), lead with a "detect first" step, then route to the matching reference. Example: a framework skill opens by reading composer.json/package.json to find the installed major version, then loads references/<name>-vN.md. This is what keeps a skill correct as the tool evolves.
  • Separate the stable layer from the volatile layer. Version-independent principles live in SKILL.md; exact namespaces/signatures live in per-version references (and say "verify against the official docs for this version" rather than freezing volatile API).
  • Layout for a multi-file skill:
    skills/<name>/
      SKILL.md                 # detect + route + stable principles
      references/<topic>.md     # loaded on demand
      references/upgrades/*.md   # e.g. migration/upgrade paths

4. Write for the agent, not a human reader

  • Imperative and concise. Every line should change what the agent does. If an explanation is longer than the thing it explains, cut it.
  • Concrete, labeled code examples beat prose. Label each example with the version/context it applies to.
  • Give a short "Output expectations" section (what the agent should produce/state).
  • No marketing, no filler, no restating the obvious.

5. Verify every fact before you write it (non-negotiable)

Skills are trusted and reused, so a wrong claim scales. Web-verify against primary/official sources (docs, changelogs, the framework's repo) before writing - do not rely on training memory for version numbers, API names, signatures, or dates. For a multi-part topic (e.g. 5 versions), run parallel research agents, one per part, each returning sourced findings, then synthesize. Cite source URLs in the reference files. Flag anything you couldn't confirm as uncertain rather than stating it confidently.

6. Test it

Install it and try it on a real task: npx skills add <owner/repo> --skill <name> (or point the agent at the local file). Confirm the description makes it activate at the right moment, the routing loads the right reference, and the guidance produces correct output. Iterate on the description first if it doesn't fire.

7. Publish (Majd's workflow)

Skills live in the monorepo majdghithan/agent-skills (local: /Users/user/Desktop/Code/agent-skills).

  1. Create skills/<name>/SKILL.md (+ references/ if multi-file).
  2. Add a row to the repo README.md skills table.
  3. Commit as Majd Ghithan <majd.ghithan20@gmail.com> (never the session email; no AI co-author trailer), then push:
    git add -A && git commit -m "Add <name> skill" && git push
  4. Installable immediately: npx skills add majdghithan/agent-skills --skill <name>. skills.sh and claudemarketplaces.com auto-index the public repo (no submit form). To use it everywhere, install globally: add -g.

Checklist

  • Gap confirmed (npx skills find - not already well-covered)
  • name kebab-case = folder; description loaded with triggers + when-to-use + when-not
  • SKILL.md lean; detail in references/; detect-first routing if there are variants
  • Stable principles separated from volatile per-version syntax
  • Every factual claim web-verified against official sources; sources cited; uncertainties flagged
  • Concrete, version-labeled examples
  • Tested by installing and running on a real task
  • Published to majdghithan/agent-skills (README row, correct commit identity, pushed)

Sources

Anthropic - Agent Skills · skills.sh CLI

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.