agentleFS
edrethardoedrethardo/werkbank-board/.claude/skills/creating-skills/SKILL.md

creating-skills

Skill0 starsChanged 3 months agoMITView on GitHub

Use when a task type has occurred twice, when the user says "always do it this way", or when following a skill produced a bad result — extract or improve a skill.

31 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use building-chatgpt-apps, building-chat-interfaces 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.
  • mjunaidcaSkill

    building-chatgpt-apps

    Guides creation of ChatGPT Apps with interactive widgets using OpenAI Apps SDK and MCP servers. Use when building ChatGPT custom apps with visual UI components, embedded widgets, or rich interactive experiences. Covers widget architecture, MCP server setup with FastMCP, response metadata, and Developer Mode configuration. NOT when building standard MCP servers without widgets (use building-mcp-servers skill instead).

    Also in 4 projects that use this

  • mjunaidcaSkill

    building-chat-interfaces

    Build AI chat interfaces with custom backends, authentication, and context injection. Use when integrating chat UI with AI agents, adding auth to chat, injecting user/page context, or implementing httpOnly cookie proxies. Covers ChatKitServer, useChatKit, and MCP auth patterns. NOT when building simple chatbots without persistence or custom agent integration.

    Also in 4 projects that use this

  • +10 more from mjunaidca

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 28 edited versions we read keep these.

  • 0

    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. Very specific:

    Every skill of this tool lives in this repo's .claude/skills/.

    line 32
  2. Very specific:

    If the improvement is generic — it would help ANY project using this setup, not just this tool — also offer to carry it back to the kit: …

    line 37

“Common” is measured: kept by at least 6 in 10 of the 28 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

Creating Skills

Recurring work becomes a skill; failing skills get improved. This is how the system learns — without it, every session starts from zero.

When

  • The same shape of task has happened twice → skill, named after the task (import-monthly-csv, generate-quarterly-report).
  • The user states a standing preference ("always format dates like this") → skill, or a line added to the existing skill that owns that territory.
  • Following a skill produced a bad result → improve THAT skill now: fix the instruction, bump version, journal what went wrong and what changed (journaling skill).

Writing one

  • A folder with a SKILL.md; frontmatter: name, description, version (plain integer, bump on every change).
  • The description is a TRIGGER, not a topic: "Use when <concrete situation>". A skill that never fires is dead weight — write the description for the moment of need.
  • Short, imperative steps. Reference material goes in extra files in the skill folder, loaded only when needed.

Where a skill lives

Every skill of this tool lives in this repo's .claude/skills/. Nothing is installed machine-wide: skills here load in this project and nowhere else, so improving one can never change how Claude behaves in the user's other projects. Edit the skill in place and bump its version:.

If the improvement is generic — it would help ANY project using this setup, not just this tool — also offer to carry it back to the kit: the project's own kit handshake file (if it has one) records kit_origin. If that is a local folder, apply the same change and version bump there and journal it in both places. If it is only a URL, or the user does not want to bother, journal the improvement here with a "not carried back" note and move on; the tool keeps working either way.

Proven patterns

  • One create-X skill per recurring content type — a family of small, named skills beats one generic "do stuff" skill.
  • Mature tools: extract <tool>-reference / <tool>-how-to project skills from overgrown docs (see documenting), marked "extracted from X — keep in sync".

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.