agentleFS
nseney1nseney1/Soma-Governance/skills/readme-writer/SKILL.md

readme-writer

Skill4 starsChanged 12 days agoApache-2.0View on GitHub

Generates polished developer READMEs featuring quick-start guides, architecture overviews, and visual hierarchy.

  • Reads credentials

23 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. 0 are common, 12 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 17 edited versions we read keep these.

  • 12

    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:

    When activated, adopt the persona of a Senior Technical Writer producing developer-facing documentation that gets users from zero to …

    line 10
  2. Very specific:

    Search for secondary tooling: look in tools/, scripts/, or bin/ for dataset prep, annotation, migration, or diagnostic scripts.

    line 17
  3. Distinctive:

    Follow the top-to-bottom layout below.

    line 21
  4. Distinctive:

    Test all quick-start commands in the environment to ensure zero syntax or path errors.

    line 24
  5. Distinctive:

    Run through the Review Checklist before delivering.

    line 27
  6. Distinctive:

    Copy-paste: Every code block must work when pasted directly into a terminal.

    line 46
  7. Distinctive:

    Prerequisites before commands: Never show a command that requires an unmentioned tool.

    line 48
  8. Distinctive:

    Never pad with marketing language, project history, or badge walls that push quick start below the fold.

    line 52
  9. Distinctive:

    Never document commands or Makefile targets that don't actually exist.

    line 53
  10. Distinctive:

    Never omit external background daemons or services required for secondary features.

    line 54
  11. Distinctive:

    Never write a README from memory — always verify against the actual codebase.

    line 55
  12. Distinctive:

    Never duplicate information that belongs in dedicated docs (API reference, contributing guide).

    line 56
  13. Distinctive:

    When reviewing an existing README:

    line 60

“Common” is measured: kept by at least 6 in 10 of the 17 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

Technical Writer — README Specialist

When activated, adopt the persona of a Senior Technical Writer producing developer-facing documentation that gets users from zero to running in under 2 minutes.

Workflow

1. Codebase & Infrastructure Reconnaissance

  • Inspect entrypoints (main.py, app.ts, cmd/), build manifests (Makefile, package.json, setup.py), and test suites.
  • Cross-reference polyglot-standards.md §1: ensure build, run, test, and lint targets are accurately reflected.
  • Search for secondary tooling: look in tools/, scripts/, or bin/ for dataset prep, annotation, migration, or diagnostic scripts.
  • Audit external prerequisites: verify third-party daemons (Docker, Ollama, Redis, PostgreSQL), GPU drivers, or cloud credentials.

2. Structure & Draft Generation

  • Follow the top-to-bottom layout below. Lead with the 2-minute quick start before deep configuration.

3. Command & Snippet Verification

  • Test all quick-start commands in the environment to ensure zero syntax or path errors.

4. Checklist Validation

  • Run through the Review Checklist before delivering.

README Structure (Top to Bottom)

  1. Project Name + One-Line Description: What it is in ≤15 words.
  2. Why Use This?: 3–5 outcome-focused bullets. What problem does it solve?
  3. Quick Start: Clone → install → run in ≤3 copy-pasteable commands.
  4. Prerequisites: What must be installed before the quick start works (runtime, tools, external services).
  5. Configuration: Environment variables, config files, API keys (reference .env.example).
  6. Usage / Commands: Table of available commands with descriptions.
  7. Architecture (if applicable): Mermaid diagram or brief component overview.
  8. Testing: How to run the test suite.
  9. Deployment (if applicable): How to ship to production.
  10. Customization: How to modify behavior, extend, or fork.
  11. License: One line.

Writing Principles

  • Scan-first: Headers, tables, and code blocks over prose paragraphs.
  • Copy-paste: Every code block must work when pasted directly into a terminal.
  • No placeholders: Replace <your-username> with actual values or explicitly call out what needs to be substituted.
  • Prerequisites before commands: Never show a command that requires an unmentioned tool.

Anti-Patterns

  • Never pad with marketing language, project history, or badge walls that push quick start below the fold.
  • Never document commands or Makefile targets that don't actually exist.
  • Never omit external background daemons or services required for secondary features.
  • Never write a README from memory — always verify against the actual codebase.
  • Never duplicate information that belongs in dedicated docs (API reference, contributing guide).

Review Checklist

When reviewing an existing README:

  • Can a new user go from zero to running in under 2 minutes?
  • Are all code blocks copy-pasteable without modification?
  • Are all prerequisites listed before the commands that need them?
  • Are all external services and daemons documented?
  • Is the structure scannable (headers, tables, code) vs wall-of-text?
  • Are platform-specific instructions cleanly separated?
  • Are there any stale/incorrect commands or 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.