agentleFS
jzhao234jzhao234/big-brain/template/CLAUDE.md

big-brain / template

CLAUDE.md1 starsChanged 31 days agoMITView on GitHub

You are working inside the user's second brain, a big-brain markdown vault. Treat it as their external memory: read it before answering questions about their work, and write to it as things happen, not at the end. At the start of a session that touches ongoing work, get the lay of the land first: call the brain_overview MCP tool (or run big-brain status). Don't guess what the user is working on when the vault can tell you. Write notes the…

The file

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

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    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. Very specific:

    Prefer append_note over rewriting.

    line 22
  2. Distinctive:

    Write notes the user will still understand in a year: complete sentences, context included, no conversation-specific shorthand.

    line 27

≈ “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

Agent instructions for this vault

You are working inside the user's second brain, a big-brain markdown vault. Treat it as their external memory: read it before answering questions about their work, and write to it as things happen, not at the end.

Orientation

At the start of a session that touches ongoing work, get the lay of the land first: call the brain_overview MCP tool (or run big-brain status). Don't guess what the user is working on when the vault can tell you.

When to write

  • Something worth keeping, unclear where it goes → capture to inbox. Never let information die in the conversation.
  • A decision was made, progress happened, something broke → one line via daily_log, and to the relevant project's Log section if it belongs to a project.
  • New commitment with multiple steps → create_project (status active, a clear Goal).
  • New action item → add_task on the project it belongs to; only use the daily note for one-off errands.
  • Learned something durable (how a system works, a config that matters, a person's role) → a note in notes/, reference/, or people/ with [[wikilinks]] to related notes.

Conventions

  • Wikilinks ([[Note Title]]) are the connective tissue. Link every new note to at least one existing note or project.
  • Frontmatter drives structure: type, status, tags, due. Update status with set_project_status / update_frontmatter rather than editing prose.
  • Tasks are - [ ] checkboxes. Metadata: 📅 YYYY-MM-DD due, ⏫ high priority, ✅ completion date (added automatically by complete_task). To reschedule, reprioritize, reword, reopen, or cancel (- [-]) a task, use update_task rather than rewriting the note.
  • Prefer append_note over rewriting; never delete. Use archive_note instead.
  • Dates are absolute (2026-07-07), never "yesterday" or "next week".

Tone of vault content

Write notes the user will still understand in a year: complete sentences, context included, no conversation-specific shorthand. A note should stand alone.

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.