agentleFS
Sapere-IASapere-IA/sdd-onboarding-kit/AGENTS.md

sdd-onboarding-kit

AGENTS.md19 starsChanged 37 days agoMITView on GitHub

This repository is an SDD onboarding kit. It is not the target project configuration. These instructions apply to any coding agent (Claude Code, Codex, Cursor, OpenCode, Antigravity or another harness) opened in this repository. When the developer asks to install or configure SDD in another repository: The final goal is to create a project-specific SDD harness, not to explain SDD abstractly. When the developer asks to make changes in this repository:

The file

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

10 rules. About 0 are common, 5 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When the developer asks to install or configure SDD in another repository:

    line 5
  2. Very specific:

    Start by reading instructions.md.

    line 7
  3. Very specific:

    Use the files under templates/, agents/, skills/, hooks/, mcps/, and scripts/ as source material.

    line 10
  4. Distinctive:

    Do not treat this kit's root AGENTS.md/CLAUDE.md as the final project instruction file.

    line 11
  5. Distinctive:

    Ask the developer before enabling hooks, configuring MCPs, changing git policy, or assuming test/lint commands.

    line 14
  6. Distinctive:

    When the developer asks to make changes in this repository:

    line 18
  7. Very specific:

    If todolist.md exists (it is gitignored and local-only), read it: it contains the tasks to be done.

    line 20
  8. Very specific:

    If bugs.md exists (also gitignored and local-only), check that the bugs it describes are solved.

    line 22
  9. Very specific:

    After adding, renaming or deleting kit files, run scripts/update-manifest.sh to regenerate manifest.md.

    line 24
  10. Distinctive:

    Keep every installable file harness-neutral: no harness name or harness-specific command in agents, skills, templates or policies unless it …

    line 25

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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

Instructions for coding agents working inside this kit

This repository is an SDD onboarding kit. It is not the target project configuration. These instructions apply to any coding agent (Claude Code, Codex, Cursor, OpenCode, Antigravity or another harness) opened in this repository.

When the developer asks to install or configure SDD in another repository:

  1. Start by reading instructions.md.
  2. Then read questions.md.
  3. Then read output-project-structure.md.
  4. Use the files under templates/, agents/, skills/, hooks/, mcps/, and scripts/ as source material.
  5. Do not treat this kit's root AGENTS.md/CLAUDE.md as the final project instruction file.
  6. Generate or update the target project's own AGENTS.md from templates/AGENTS.md.template (plus the CLAUDE.md import stub from templates/CLAUDE.md.template when Claude Code is one of the harnesses).
  7. Map the kit's reference layout (.claude/) to the harness being installed using reference/harness-primitives.md.
  8. Ask the developer before enabling hooks, configuring MCPs, changing git policy, or assuming test/lint commands.

The final goal is to create a project-specific SDD harness, not to explain SDD abstractly.

When the developer asks to make changes in this repository:

  1. If todolist.md exists (it is gitignored and local-only), read it: it contains the tasks to be done. If it does not exist, ask the developer what to change.
  2. Complete the tasks. You may need to modify some of the hooks, MCPs, templates, etc.
  3. If bugs.md exists (also gitignored and local-only), check that the bugs it describes are solved.
  4. Mark with X all the checkboxes of the tasks/bugs you complete.
  5. After adding, renaming or deleting kit files, run scripts/update-manifest.sh to regenerate manifest.md.
  6. Keep every installable file harness-neutral: no harness name or harness-specific command in agents, skills, templates or policies unless it is inside a per-harness table or example.

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.