agentleFS
DGouronDGouron/review-flow/.claude/skills/skill-creator/SKILL.md

skill-creator

Skill44 starsChanged 6 months agoMITView on GitHub

Guide for creating or updating a ReviewFlow skill. Use when the user wants to create a new skill, update an existing skill, or asks how to add a slash command that extends Claude's capabilities in this project.

189 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use taskflow-inbox-triage, mermaid-diagram-generator 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.
  • AtomicBot-aiSkill

    taskflow-inbox-triage

    Also in 6 projects that use this

  • shalombSkill

    mermaid-diagram-generator

    Guide on creating and editing Mermaid charts safely. Use when asked to draw a diagram, flowchart, or architecture chart. Triggers on: 'mermaid', 'diagram', 'flowchart', 'visualize'.

    Also in 5 projects that use this

  • naman10parikhSkill

    ecosystem-reference

    Quick-access directory of Claude Code ecosystem repos, skills, plugins, MCPs, and patterns. Use when improving the harness.

    Also in 5 projects that use this

  • naman10parikhSkill

    fractal-delegation

    Also in 5 projects that use this

  • Liang-MHSkill

    stock_analysis

    获取 A 股、基金和港股的实时行情和技术分析数据

    Also in 5 projects that use this

  • nicepkgSkill

    community-led-growth

    Expert in community-led growth (CLG) - leveraging user communities to drive acquisition, retention, and expansion. Covers building developer communities, user groups, ambassador programs, and turning customers into advocates. Knows the difference between community as a feature vs community as a growth engine, and how to measure community ROI. Use when "community-led, CLG, ambassador program, champion program, user community, developer community, word of mouth, user group, " mentioned.

    Also in 5 projects that use this

  • +6 more from naman10parikh

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 185 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write the body — imperative instructions, low prose

    line 47
  2. Very specific:

    Test it: invoke /skill-name and verify Claude follows the workflow

    line 48
  3. Distinctive:

    Write in English — no exceptions (code, body, comments)

    line 70
  4. Distinctive:

    Use imperative form: "Run", "Read", "Stop", never "You should run"

    line 71
  5. Distinctive:

    Keep the body under 400 lines

    line 73
  6. Distinctive:

    Prefer tables and bullet lists over prose paragraphs

    line 75
  7. Distinctive:

    After creating a skill, add it to the skills table in CLAUDE.md:

    line 99

“Common” is measured: kept by at least 6 in 10 of the 185 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 Creator

Guide for creating effective skills in the ReviewFlow project.

About Skills

Skills are modular, self-contained packages that extend Claude's capabilities with specialized workflows, domain knowledge, and tool integrations. They transform Claude from a general-purpose agent into a project-specialized one.

What Skills Provide

  1. Specialized workflows — Multi-step procedures (e.g. /ship, /tdd, /implement-feature)
  2. Domain expertise — ReviewFlow-specific knowledge, architecture patterns, business rules
  3. Bundled resources — References and assets loaded on demand

Core Principles

Concise is Key

The context window is a shared resource. Only add context Claude does not already have. Challenge each paragraph: "Does Claude really need this?" If not, remove it.

Set Appropriate Degrees of Freedom

  • High freedom (text instructions): multiple approaches valid, context-dependent
  • Medium freedom (pseudocode/templates): preferred pattern exists, variation acceptable
  • Low freedom (shell scripts): operations are fragile, consistency critical

Anatomy of a ReviewFlow Skill

.claude/skills/<skill-name>/
├── SKILL.md            (required)
│   ├── YAML frontmatter (required: name + description)
│   └── Markdown instructions
└── references/         (optional — bundled docs loaded by reference)

Skill Creation Process

  1. Understand the skill with 2-3 concrete usage examples
  2. Draft the frontmatter (name, description, user-invocable)
  3. Write the body — imperative instructions, low prose
  4. Test it: invoke /skill-name and verify Claude follows the workflow
  5. Register in CLAUDE.md skills table

Frontmatter Reference

---
name: skill-name            # kebab-case, matches folder name
description: >              # PRIMARY TRIGGER — LLM reads only this initially.
  One or two sentences.     # Include natural-language trigger phrases.
  "Use when the user says X, Y, Z."
user-invocable: true        # true = can be invoked with /skill-name
allowed-tools: Bash, Read   # optional — restrict available tools
---

Official fields: name, description, user-invocable, allowed-tools, disable-model-invocation, context.

triggers: [] is NOT an official field — it is silently ignored. The description is the only trigger mechanism.

Writing Guidelines

  • Write in English — no exceptions (code, body, comments)
  • Use imperative form: "Run", "Read", "Stop", never "You should run"
  • description is the primary trigger — include natural trigger phrases in it
  • Keep the body under 400 lines
  • Include only what Claude does not already know
  • Prefer tables and bullet lists over prose paragraphs
  • Match ReviewFlow conventions (see CLAUDE.md and .claude/rules/coding-standards.md)

What NOT to Include

  • Generic programming knowledge Claude already has
  • README-style context that belongs in docs
  • Comments explaining what the skill is (the frontmatter does that)
  • Skills are for AI agents, not humans — do not write tutorials

ReviewFlow-Specific Conventions

ConventionDetail
Package manageryarn — never pnpm or npm
Main branchmaster — never main
Quality gateyarn verify (typecheck + lint + test:ci)
Commit formatConventional Commits via commitlint
LanguageEnglish everywhere (code, tests, commits, logs)
Imports@/ alias + .js extension mandatory
No barrel exportsDirect imports only, no index.ts re-exports

Skill Registration

After creating a skill, add it to the skills table in CLAUDE.md:

| `/skill-name` | When to use this skill |

Quick Template

---
name: my-skill
description: Brief description. Use when the user says "X", "Y", or wants to Z.
user-invocable: true
---

# My Skill — Short Title

## Activation

This skill activates with `/my-skill`.

## Workflow

### Step 1: <Action>

<Instructions>

### Step 2: <Action>

<Instructions>

## Rules

- NEVER do X
- ALWAYS do Y

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.