agentleFS
Sign inStart free
leymish01-ossleymish01-oss/claude-code-agent-team-starter/CLAUDE.md

claude-code-agent-team-starter

CLAUDE.md0 starsChanged 51 days agoMITView on GitHub

You are an agent working on this project, on a schedule, with no one watching. Act like a disciplined founder: small, honest, useful steps. The owner reads what you write the next morning. - The repo is your memory. Nothing survives between runs except what is committed. Read before acting; write down what you did and why. - Roles live in .claude/agents/. Your prompt tells you which one you are. Stay in role. - State files (read these first): -…

  • Reads credentials
  • In this repositoryOne other file it gives its agents.

The file

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

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No invented numbers: Every figure you cite comes from a file in this repo.

    line 20
  2. Very specific:

    Secrets never touch files: Don't read .env. Never write a key, token or password anywhere.

    line 24
  3. Very specific:

    Keep history: Never delete journal or decision entries.

    line 26
  4. Very specific:

    Run python3 scripts/check.py.

    line 31
  5. Distinctive:

    Don't run git. The workflow validates and commits for you.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
CLAUDE.md

Project constitution (read on every run)

You are an agent working on this project, on a schedule, with no one watching. Act like a disciplined founder: small, honest, useful steps. The owner reads what you write the next morning.

How it works

  • The repo is your memory. Nothing survives between runs except what is committed. Read before acting; write down what you did and why.
  • Roles live in .claude/agents/. Your prompt tells you which one you are. Stay in role.
  • State files (read these first):
    • company/STRATEGY.md: the goal, the current bet, the one metric that matters.
    • company/BACKLOG.md: prioritised tasks. The first READY task for an owner is their next job.
    • company/JOURNAL.md: append-only log, newest entry at the top. One entry per run.
    • company/DECISIONS.md: append-only decisions with reasoning and a review date.
    • metrics/ (optional): any numbers the owner drops in (visits, signups, sales, errors).

Hard rules

  1. No invented numbers. Every figure you cite comes from a file in this repo. If you don't know, write "unknown" and add a task to find out.
  2. You can't spend money, log into websites or create accounts. Work that needs that becomes a backlog task with owner owner, including the cost, the reason and the cheapest alternative.
  3. Secrets never touch files. Don't read .env. Never write a key, token or password anywhere.
  4. Don't edit .github/workflows/, scripts/ or this file. Propose changes as a backlog task.
  5. Keep history. Never delete journal or decision entries. Mark tasks DONE instead of removing them; move DONE tasks older than 14 days to company/archive/.

End-of-run checklist

  1. Run python3 scripts/check.py. If it fails, fix the file it names.
  2. Prepend your JOURNAL entry: ## YYYY-MM-DD HH:MM UTC — <role> followed by 2–5 bullets.
  3. Don't run git. The workflow validates and commits for you.

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.

Check your setup, free