agentleFS
kunalsurikunalsuri/ai-fication-kit/CLAUDE.md

ai-fication-kit

CLAUDE.md3 starsChanged 4 months agoApache-2.0View on GitHub

A Toolkit to Give AI Coding Agents a Trusted Map of Any Existing/Legacy Repo Stack: JavaScript/TypeScript. If CLAUDE_bkp_*.md exists, it is the prior configuration. Knowledge was extracted from it during /cold-start — do not re-read it; use ai/guide/ instead.

The file

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

11 rules. About 0 are common, 4 are distinctive, and 7 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.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run the suite matching what you changed BEFORE claiming success.

    line 15
  2. Distinctive:

    Respect existing boundaries: Treat unfamiliar, load-bearing code as frozen until the module map says otherwise.

    line 18
  3. Very specific:

    Check ai/guide/MODULE_MAP.md Stability before editing any file.

    line 19
  4. Distinctive:

    Never flip that tag yourself.

    line 21
  5. Very specific:

    No Phantom Bugs & Configuration Churn: Do not rewrite, restructure, or simplify configuration or instruction files (CLAUDE.md, AGENTS.md, …

    line 22
  6. Very specific:

    Verify claims: Before declaring a task finished, run node install.mjs verify .

    line 23
  7. Very specific:

    Record work: Every unit of work (feature, bugfix, refactor, process change) ends with a row appended to ai/lab/WORKLOG.md — even work done …

    line 24
  8. Distinctive:

    Prefer grep and line counts over whole-file reads.

    line 40
  9. Very specific:

    If CLAUDE_bkp_*.md exists, it is the prior configuration.

    line 52
  10. Very specific:

    If ai/guide/MODULE_MAP.md still has placeholder rows, run /cold-start once, then a human audits (set Stability, flip [inferred] -> …

    line 56
  11. Very specific:

    If backup files (e.g. CLAUDE_bkp_*.md) exist, the agent will extract and reuse knowledge from your prior configuration to seed the …

    line 58

≈ “Common” is an estimate: at least 1 in 5 of the 27,641 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
<!-- Copyright (c) 2026 Kunal Suri (CEA LIST). All rights reserved. -->

ai-fication-kit — Claude Code project memory

Imports tool-agnostic rules (Claude Code native directive)

@AGENTS.md

Relative link for non-Claude agents: AGENTS.md

A Toolkit to Give AI Coding Agents a Trusted Map of Any Existing/Legacy Repo Stack: JavaScript/TypeScript.

Build & test — VERIFY in the cold-start pass before trusting

  • Build: npm install
  • Test: npm test
  • Test locations: test/
  • Run the suite matching what you changed BEFORE claiming success.

Hard rules (non-negotiable)

  • Respect existing boundaries. Treat unfamiliar, load-bearing code as frozen until the module map says otherwise.
  • Check ai/guide/MODULE_MAP.md Stability before editing any file. frozen = hands off.
  • Anything you write into ai/ is [inferred] until a human flips it to [verified]. Never flip that tag yourself.
  • No Phantom Bugs & Configuration Churn: Do not rewrite, restructure, or simplify configuration or instruction files (CLAUDE.md, AGENTS.md, package.json, or anything in ai/guide/) based on quick searches or automated suggestions. Keep edits surgical. Never replace detailed guides with simplified stubs.
  • Verify claims: Before declaring a task finished, run node install.mjs verify . --strict (or your repo's verify script) to ensure no file paths in the knowledge documents are broken.
  • Record work: Every unit of work (feature, bugfix, refactor, process change) ends with a row appended to ai/lab/WORKLOG.md — even work done outside the slash commands.

Where to look — read on demand, do NOT pre-load

  • Role → path manifest: ai/INDEX.md
  • Machine-readable facts: ai/repo-profile.json (orient output; deterministic)
  • Find code by area: ai/guide/MODULE_MAP.md <- START HERE to locate anything
  • What & why: ai/guide/PROJECT_OVERVIEW.md
  • System shape: ai/guide/ARCHITECTURE.md
  • Feature -> files: ai/guide/FEATURE_MAP.md
  • Conventions: ai/guide/CONVENTIONS.md
  • Generated analysis: ai/analysis/ (feature catalog, diagrams, audit reports)
  • Specs / ADRs / evals: ai/lab/
  • Feature roadmap: ai/lab/ROADMAP.md (planned backlog + shipped index; cousin of WORKLOG)

Token discipline

  • Locate via MODULE_MAP.md, then open only needed files. Don't crawl the tree.
  • Prefer grep and line counts over whole-file reads.
  • Delegate isolated/heavy work to subagents: repo-explorer (read-only exploration), feature-builder (implements), test-runner (tests).

Repo intelligence (the ai/ knowledge-base)

The ai/ folder is the single source of truth for repository intelligence. It is tool-agnostic: any AI coding agent (Claude, Cursor, Copilot, Codex) reads the same verified maps, architecture docs, and feature catalogs. New features, refactors, and onboarding all start here. Claude Code's auto-memory is machine-local and unshared — the committed ai/ layer is the source of truth every agent and teammate reads, so record durable knowledge there.

If CLAUDE_bkp_*.md exists, it is the prior configuration. Knowledge was extracted from it during /cold-start — do not re-read it; use ai/guide/ instead.

Cold start

If ai/guide/MODULE_MAP.md still has placeholder rows, run /cold-start once, then a human audits (set Stability, flip [inferred] -> [verified]) before features are built. If backup files (e.g. CLAUDE_bkp_*.md) exist, the agent will extract and reuse knowledge from your prior configuration to seed the ai/guide/ documents.

<!-- Installed by ai-fication-kit 0.1.0 on 2026-06-17. -->

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.