agentleFS
endearqbendearqb/endearqb-skills/skills/readme-first-builder/SKILL.md

readme-first-builder

Skill19 starsChanged 6 months agoMITView on GitHub

Initialize or upgrade a project to the full README First architecture from endearqb/ReadmeFirst. Use only when the user asks to initialize README First, install the README First framework, create the complete AGENTS.md + README.md + .ai/changes + .ai/decisions structure, or retrofit a repository with the README First architecture. After initialization, normal project work should follow the target project's AGENTS.md instead of this skill.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 10 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    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:

    Use this skill as a one-time initializer or upgrader for a project-level README First architecture.

    line 10
  2. Very specific:

    Read references/readmefirst-source.md for the upstream repository and file roles.

    line 14
  3. Distinctive:

    When network access is available, inspect the latest upstream files before initializing.

    line 19
  4. Distinctive:

    If AGENTS.md is missing, create it from the upstream ReadmeFirst AGENTS.md content, then adapt only project-specific placeholders if …

    line 24
  5. Distinctive:

    If AGENTS.md already exists, merge the README First protocol into it without overwriting local rules.

    line 25
  6. Distinctive:

    Ensure the root README.md acts as the project map: purpose, setup, commands, top-level directory responsibilities, and links to important …

    line 26
  7. Very specific:

    Create .ai/changes/ and .ai/decisions/ if missing.

    line 27
  8. Distinctive:

    Add or update directory-level README.md files only for key long-lived directories.

    line 28
  9. Distinctive:

    Verify paths and commands mentioned in the generated docs.

    line 30
  10. Distinctive:

    Never overwrite an existing AGENTS.md or README.md blindly.

    line 34
  11. Distinctive:

    Treat existing project rules as local authority unless they conflict with safety or the user's explicit request.

    line 35
  12. Distinctive:

    If the upstream AGENTS.md conflicts with existing project instructions, report the conflict and choose the least destructive merge.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

README First Builder

Purpose

Use this skill as a one-time initializer or upgrader for a project-level README First architecture. Do not use it as the default way to write every README after a project already has AGENTS.md; once initialized, the target project's own AGENTS.md becomes the operating protocol.

Canonical Source

Read references/readmefirst-source.md for the upstream repository and file roles. The canonical project is endearqb/ReadmeFirst:

  • README.md: principle and architecture explanation.
  • AGENTS.md: executable agent protocol.

When network access is available, inspect the latest upstream files before initializing. If network is unavailable, explain that the initialization is based on the locally known structure and recommend a later sync.

Initialization Workflow

  1. Inspect the target project root and confirm whether AGENTS.md, README.md, and .ai/ already exist.
  2. If AGENTS.md is missing, create it from the upstream ReadmeFirst AGENTS.md content, then adapt only project-specific placeholders if necessary.
  3. If AGENTS.md already exists, merge the README First protocol into it without overwriting local rules. Keep stricter or project-specific local rules.
  4. Ensure the root README.md acts as the project map: purpose, setup, commands, top-level directory responsibilities, and links to important directory READMEs.
  5. Create .ai/changes/ and .ai/decisions/ if missing. Add lightweight starter templates only when useful.
  6. Add or update directory-level README.md files only for key long-lived directories. Do not cover generated, dependency, cache, build, log, or temporary directories.
  7. Record the initialization in .ai/changes/YYYY-MM-DD.md.
  8. Verify paths and commands mentioned in the generated docs.

Existing File Rules

  • Never overwrite an existing AGENTS.md or README.md blindly.
  • Treat existing project rules as local authority unless they conflict with safety or the user's explicit request.
  • Preserve user-authored prose and append or merge README First sections.
  • If the upstream AGENTS.md conflicts with existing project instructions, report the conflict and choose the least destructive merge.

Handoff After Initialization

End by telling future agents to follow the newly created or merged project AGENTS.md. For later ordinary code, documentation, review, or refactor tasks, use the project's own AGENTS.md and relevant directory READMEs rather than re-triggering this skill.

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.