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

9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.

• ≈0

  Common · estimate

  Many files like this have a rule of this kind.

• ≈4

  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:

   Keep sibling CLAUDE.md and AGENTS.md files aligned.

   line 7
2. Very specific:

   Treat docs/spec.md as stable product truth, tasks/current.md as a derived status snapshot, and tasks/todo.md as the deferred-goal ledger.

   line 8
3. Very specific:

   Treat tasks/lessons.md, tasks/research.md, and .ai/harness/policy.json as durable workflow context.

   line 9
4. Very specific:

   Use .ai/context/context-map.json and .ai/context/capabilities.json to discover functional-block contracts.

   line 10
5. Very specific:

   Do not infer local CLAUDE.md or AGENTS.md files from broad physical layouts such as apps/*, packages/*, or services/*.

   line 11
6. Distinctive:

   Put capability-specific ownership, entrypoints, and verification commands in explicitly selected functional-block contracts.

   line 12
7. Distinctive:

   Keep root context concise.

   line 13
8. Very specific:

   Treat _ref/ as ignored external reference material and _ops/ as ignored local operations state.

   line 14
9. Distinctive:

   Prefer repo-local workflow artifacts over tool-specific chat memory.

   line 15

≈ “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.