agentleFS
Sign inStart free
pankaj-mahaurpankaj-mahaur/Agent-Daemon/CLAUDE.md

Agent-Daemon

CLAUDE.md0 starsChanged 5 days agoMITView on GitHub

This project uses agent-daemon — a self-improving runtime for AI coding agents with multi-agent orchestration. This section is managed by ad init — re-running refreshes it. The complete agent-daemon manual — task-complexity gate, skill decision tree, daemon workflow, mid-session memory discipline, and the mandatory session-close protocol — lives in AD-INSTRUCTIONS.md at the repo root (kept out of CLAUDE.md to save per-session context). Read AD-INSTRUCTIONS.md before any substantial or specialized task — audit, implement, debug, review, migration, security, multi-agent orchestration, or…

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:

    CLI: All commands use the ad shorthand (ad doctor, ad init, ad tt, ad memory, ad review)

    line 8

≈ “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-daemon:start -->

agent-daemon

This project uses agent-daemon — a self-improving runtime for AI coding agents with multi-agent orchestration. This section is managed by ad init — re-running refreshes it.

  • Memory: .agent-daemon/memory/ — project context and learnings captured continuously by local hooks; session-close digests add richer summaries
  • Multi-agent: ad tc to create teams, ad sp to spawn workers in isolated git worktrees, ad ts for status
  • CLI: All commands use the ad shorthand (ad doctor, ad init, ad tt, ad memory, ad review)
  • Skills: Auto-triggering skills in ~/.claude/skills/ — code review, debugging, orchestration, etc.
  • Self-improvement: deterministic local capture and optional session-close digests write SQLite memory; skill outcomes remain review-gated

📖 Full operating manual → AD-INSTRUCTIONS.md

The complete agent-daemon manual — task-complexity gate, skill decision tree, daemon workflow, mid-session memory discipline, and the mandatory session-close protocol — lives in AD-INSTRUCTIONS.md at the repo root (kept out of CLAUDE.md to save per-session context).

Read AD-INSTRUCTIONS.md before any substantial or specialized task — audit, implement, debug, review, migration, security, multi-agent orchestration, or session-close. It encodes which skill to invoke and the ordering/dedup/severity logic that one-shot prompting misses.

Proportionality rule (always on): trivial requests (hey, a quick question, a one-line edit) — act normally, do NOT search for skills. Anything substantial — read AD-INSTRUCTIONS.md first, then pick the narrowest matching skill before freelancing. Routing advice is also injected automatically by the capability-route-advice hook (see ad route show).

Session close ("bye", "session khatam", "done for today", "wrapping up") is a mandatory 3-step protocol — see the Session-close section of AD-INSTRUCTIONS.md.

<!-- agent-daemon:end -->

This repo: before every push

Before any commit, push, PR, merge or release, run the ad-pre-push skill (.claude/skills/ad-pre-push/SKILL.md) without being asked: sync the docs and CHANGELOG with the code, update the skills that encode the change, record what was learned, check that nothing personal or internal is pushed, then verify (node runtime/scripts/check-doc-links.mjs, npm test).

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