agentleFS
codervisorcodervisor/leanspec/AGENTS.md

leanspec

AGENTS.md294 starsChanged 6 months agoMITView on GitHub

Lightweight spec methodology for AI-powered development. This project uses the Agent Skills framework for domain-specific guidance. Read the appropriate skill when working on related tasks. Lean-spec dogfoods its own SDD methodology with GitHub issues as the canonical spec medium. The specs/ directory is a frozen historical snapshot — new specs are GitHub issues on codervisor/leanspec. A pre-commit hook and CI job (freeze-specs) block additions/modifications under specs/; see specs/FROZEN.md.

  • Commits and pushes

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. codervisor/lean-spec★ 296100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 33,315 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository8 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈12

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use when: Starting non-trivial work, deciding spec-vs-trivial, picking branch name

    line 17
  2. Distinctive:

    issue-spec: - Create lean-spec style GitHub issue specs

    line 20
  3. Distinctive:

    Use when: Asked to "spec this" or planning work that needs alignment

    line 22
  4. Distinctive:

    Use when: About to git push, hitting merge conflicts, "ready to push"

    line 27
  5. Distinctive:

    Use when: CI is failing, review comments arrive, PR needs label flip, merge follow-through

    line 32
  6. Distinctive:

    Use when: Contributing code, running tests, publishing, CI/CD, or looking up commands

    line 39
  7. Distinctive:

    Use when: Testing web UIs, interacting with websites, filling forms, taking screenshots

    line 44
  8. Distinctive:

    Use when: Reading what we ship to users (this repo dogfoods that methodology via the four skills above)

    line 52
  9. Distinctive:

    Use pnpm: - Never npm or yarn. All package management uses pnpm.

    line 56
  10. Distinctive:

    Skills First: - Read the relevant skill file before starting work on development, specs, or publishing tasks.

    line 58
  11. Distinctive:

    Context Economy: - Keep specs under 2000 tokens.

    line 59
  12. Distinctive:

    Progressive Disclosure: - Use skills and references for detailed guidance.

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 42,283 AGENTS.md 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
AGENTS.md

AI Agent Instructions

Project: LeanSpec

Lightweight spec methodology for AI-powered development.

Skills

This project uses the Agent Skills framework for domain-specific guidance. Read the appropriate skill when working on related tasks.

Spec-driven development (GitHub-issue specs)

Lean-spec dogfoods its own SDD methodology with GitHub issues as the canonical spec medium. The specs/ directory is a frozen historical snapshot — new specs are GitHub issues on codervisor/leanspec. A pre-commit hook and CI job (freeze-specs) block additions/modifications under specs/; see specs/FROZEN.md.

  1. leanspec-dev-process - End-to-end SDD loop (spec → branch → PR → merge)
    • Location: .agents/skills/leanspec-dev-process/SKILL.md
    • Use when: Starting non-trivial work, deciding spec-vs-trivial, picking branch name
    • Key: No PR without a spec issue or trivial label
  2. issue-spec - Create lean-spec style GitHub issue specs
    • Location: .agents/skills/issue-spec/SKILL.md
    • Use when: Asked to "spec this" or planning work that needs alignment
    • Key: Body under ~2000 tokens; Overview / Design / Plan / Test / Provider impact / Alignment / Notes
  3. leanspec-pre-push - Pre-push checks (sync main, typecheck, clippy, spec-link)
    • Location: .agents/skills/leanspec-pre-push/SKILL.md
    • Use when: About to git push, hitting merge conflicts, "ready to push"
    • Key: CI tests a merge preview, not the branch alone — always sync main first
  4. leanspec-pr-lifecycle - Manage PR post-push (CI triage, review, label flips, merge)
    • Location: .agents/skills/leanspec-pr-lifecycle/SKILL.md
    • Use when: CI is failing, review comments arrive, PR needs label flip, merge follow-through
    • Key: Use mcp__github__* for issue/PR mutation; never gh CLI for that

Development & tooling

  1. leanspec-development - Development, commands, publishing, CI/CD, and runner research
    • Location: .agents/skills/leanspec-development/SKILL.md
    • Use when: Contributing code, running tests, publishing, CI/CD, or looking up commands
    • Key: Always use pnpm, follow DRY principle, i18n parity is mandatory
  2. agent-browser - Browser automation for testing web apps
    • Location: .agents/skills/agent-browser/SKILL.md
    • Use when: Testing web UIs, interacting with websites, filling forms, taking screenshots
    • Key: Use agent-browser CLI instead of Playwright MCP for browser automation

Reference (external)

  • leanspec (external) - Spec-Driven Development methodology for downstream users
    • Location: skills/leanspec/SKILL.md (shipped from this repo)
    • Install: npx skills add codervisor/leanspec@leanspec
    • Use when: Reading what we ship to users (this repo dogfoods that methodology via the four skills above)

Project-Specific Rules

  1. Use pnpm - Never npm or yarn. All package management uses pnpm.
  2. DRY Principle - Extract shared logic; avoid duplication.
  3. Skills First - Read the relevant skill file before starting work on development, specs, or publishing tasks.
  4. Context Economy - Keep specs under 2000 tokens. Split large tasks.
  5. Progressive Disclosure - Use skills and references for detailed guidance.
  6. GitHub issues are canonical specs - New specs are GitHub issues, not files. The specs/ directory is a frozen snapshot of pre-migration work; do not author new specs/NNN-slug/ directories. See issue-spec and leanspec-dev-process.

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.