agentleFS
Sign inStart free
Harish-hereHarish-here/maddog/CLAUDE.md

maddog

CLAUDE.md3 starsChanged 32 days agoMITView on GitHub

Beliefs: PHILOSOPHY.md. Visual identity and user-facing wording: DESIGN.md. Routing and the shipped catalog: README.md. How a change is validated — nothing compiles; model-driven tests live in tests/: CONTRIBUTING.md §Validation. Merging to main publishes: the marketplace installs from main HEAD. Run the release skill before merging anything to main — the table in its section 3 says which changes need an independent reviewer. Load-bearing instruction text goes through the section-by-section skill. The release reviewer is its only independent check, and only…

The file

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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:

    After context compaction in an advisor-mode session, re-read skills/advisor-mode/SKILL.md before the next dispatch.

    line 47

≈ “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

CLAUDE.md

Beliefs: PHILOSOPHY.md. Visual identity and user-facing wording: DESIGN.md. Routing and the shipped catalog: README.md. How a change is validated — nothing compiles; model-driven tests live in tests/: CONTRIBUTING.md §Validation.

Publishing

Merging to main publishes: the marketplace installs from main HEAD. Run the release skill before merging anything to main — the table in its section 3 says which changes need an independent reviewer. Load-bearing instruction text goes through the section-by-section skill. The release reviewer is its only independent check, and only for paths its section 3 table marks required.

Invariants

  • executor-lead and executor-judge hold no Write/Edit, and scripts/hooks/executor-guard.sh denies their Bash write-forms too. Granting either tool dissolves the invariant.
  • Adapter set — the only paths where runtime mechanics may live: agent and skill frontmatter, hooks/, scripts/, workflows/, tests/, .github/, .claude/, .claude-plugin/. Within tests/, only code under tests/harness/runtimes/ names runtime identifiers, and tests/guard/, which tests the hook scripts against their runtime's own protocol; tests/README.md and tests/requirements.txt may name the runtime they document. Shipped bodies (agents/*.md, skills/**) name capabilities (write, edit, shell, web), never runtime tool identifiers, settings keys, or APIs (PHILOSOPHY.md point 5).
  • .claude/ is repo-internal and is never registered as a plugin surface.
  • Model-driven tests score from recorded tool calls, never a grading model, and never a model's words — except, by user waiver 2026-10-08, exact matches in tests/agents/ on a fast-tier agent's PATTERNS: line and on Fast-Read's returned quotes and verdict labels. tests/harness/ enforces isolation, voided runs, and the main-branch baseline: change the harness, never bypass it.

Distribution mechanics

A plugin install auto-registers agents/, skills/, commands/, hooks/, and .mcp.json; workflows/ and scripts/ ship in the tarball but are not auto-wired (undocumented). Agent frontmatter hooks: and permissionMode: are ignored in plugin agents — guard hooks arrive via hooks/hooks.json. Skill edits take effect immediately; agent edits need /reload-plugins or a restart; workflow reload is undocumented, so restart to be sure.

Advisor sessions

After context compaction in an advisor-mode session, re-read skills/advisor-mode/SKILL.md before the next dispatch.

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