agentleFS
Sign inStart free
rust-worksrust-works/succinctly/.agents/skills/skill-writing/SKILL.md

skill-writing

Skill52 starsChanged 26 days agoMITView on GitHub

Write or improve Codex skills for this repository. Use when creating or editing a SKILL.md file or deciding whether guidance belongs in a skill.

We split this skill into the rules it gives an agent

8 rules. 0 are common, 5 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 5

    Distinctive

    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. Distinctive:

    Put instructions that apply to every task in AGENTS.md.

    line 10
  2. Very specific:

    Put task-specific workflows and reference material in .agents/skills/<name>/SKILL.md.

    line 11
  3. Distinctive:

    Keep each skill focused on one job.

    line 12
  4. Very specific:

    Create a directory with a SKILL.md file.

    line 16
  5. Distinctive:

    Write the description so Codex can select the skill from the user's request.

    line 29
  6. Very specific:

    Use agents/openai.yaml for optional display metadata, invocation policy, and tool dependencies.

    line 42
  7. Distinctive:

    Check all referenced files and links.

    line 47
  8. Distinctive:

    Keep examples aligned with tools and paths that actually exist in this repository.

    line 49

“Common” is measured: kept by at least 6 in 10 of the 5 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

Writing Codex Skills

Choose the right home

  • Put instructions that apply to every task in AGENTS.md.
  • Put task-specific workflows and reference material in .agents/skills/<name>/SKILL.md.
  • Keep each skill focused on one job. Move detailed examples or data to supporting files in the same skill directory.

Required structure

Create a directory with a SKILL.md file. Include name and description in YAML frontmatter:

---
name: example-skill
description: What this skill does and when Codex should use it.
---

# Example Skill

Follow these steps when the task matches the description.

Write the description so Codex can select the skill from the user's request. State both the task and the trigger. Write instructions as concrete actions with clear inputs and outputs.

Optional files

example-skill/
├── SKILL.md
├── references/
├── scripts/
├── assets/
└── agents/openai.yaml

Use agents/openai.yaml for optional display metadata, invocation policy, and tool dependencies. To require explicit invocation, set policy.allow_implicit_invocation: false there.

Check the result

  1. Confirm the manifest is named exactly SKILL.md and has both required frontmatter fields.
  2. Check all referenced files and links.
  3. Try a prompt that should trigger the skill and one that should not.
  4. Keep examples aligned with tools and paths that actually exist in this repository.

See the official Codex skill documentation for current fields and discovery paths.

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