agentleFS
Sign inStart free
diz5diz5/agentic-workflow-toolkit/CLAUDE.md

agentic-workflow-toolkit

CLAUDE.md1 starsChanged 3 months agoMITView on GitHub

Hard floor rules for EVERY session in this workspace. They apply before and above any skill, and they exist because long-running sessions have repeatedly drifted on exactly these points. 1. Re-read the playbook before analyzing. Before the FIRST job analysis of this session, Read .claude/skills/analyze-job/SKILL.md from disk — it changes often, and the copy in a resumed session's context is stale. Stale playbooks are the #1 cause of protocol violations. 2. Job-analysis output is Chinese (中文). Analyses, batch reports, verdicts,…

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. Very specific:

    Re-read the playbook before analyzing: Before the FIRST job analysis of this session, Read .claude/skills/analyze-job/SKILL.md from disk — …

    line 6
  2. Very specific:

    Read .claude/skills/analyze-job/SKILL.md from disk — it changes often, and the copy in a resumed session's context is stale.

    line 7
  3. Very specific:

    Every board/buffer mutation goes through python3 .claude/skills/analyze-job/scoreboard.py.

    line 13
  4. Distinctive:

    No placeholder/PENDING rows, ever.

    line 14
  5. Distinctive:

    Never improvise tables, per-job commentary, or your own summary format.

    line 17

≈ “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 instructions — agentic-workflow-toolkit

Hard floor rules for EVERY session in this workspace. They apply before and above any skill, and they exist because long-running sessions have repeatedly drifted on exactly these points.

  1. Re-read the playbook before analyzing. Before the FIRST job analysis of this session, Read .claude/skills/analyze-job/SKILL.md from disk — it changes often, and the copy in a resumed session's context is stale. Stale playbooks are the #1 cause of protocol violations.
  2. Job-analysis output is Chinese (中文). Analyses, batch reports, verdicts, status summaries — all 中文. English only inside application-answer text (submittable) and quoted material. An English wrap-up is a bug: stop and rewrite.
  3. Never touch job-scoreboard.md or job-analyses/_pending.md with Edit/Write tools. Every board/buffer mutation goes through python3 .claude/skills/analyze-job/scoreboard.py. No placeholder/PENDING rows, ever.
  4. Batch analyses follow SKILL.md Mode C exactly: progress-counter lines only while analyzers run; when all are back, one batch_summary.py --full call relayed verbatim. Never improvise tables, per-job commentary, or your own summary format. After the script output, write AT MOST one short single-clause line (the top pick). NEVER a multi-point 判断/总结 paragraph — long CJK prose corrupts on the terminal (words drop, lines merge), and it just repeats what the script already printed.
  5. Personal data never enters git. publish_guard.py + .gitignore enforce the split; anything named *.local.md, résumés, the board, and job-analyses/ stay local. When in doubt, don't commit.

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