agentleFS
black12-agblack12-ag/claude-skill/CLAUDE.md

claude-skill

CLAUDE.md0 starsChanged 4 months agoMITView on GitHub

CRITICAL RULE: You MUST check for relevant "superpowers" skills in ~/.agents/skills/ BEFORE taking any action. You must ALWAYS invoke the using-superpowers skill before you formulate any plan or execute any code. This applies to all workspaces. No exceptions. Every project may have a graphify knowledge graph at graphify-out/. RULES — ALL AI AGENTS MUST FOLLOW: - Before ANY codebase question or architecture task: read graphify-out/GRAPH_REPORT.md — god nodes and community clusters are the fastest way to understand the codebase -…

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

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

  • 11

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every project may have a graphify knowledge graph at graphify-out/.

    line 14
  2. Very specific:

    Before ANY codebase question or architecture task: read graphify-out/GRAPH_REPORT.md — god nodes and community clusters are the fastest way …

    line 17
  3. Very specific:

    If graphify-out/wiki/index.md exists, navigate it instead of reading raw files

    line 18
  4. Very specific:

    After modifying ANY code file: run graphify update . to keep the graph current (fast, AST-only, no LLM/API cost)

    line 19
  5. Very specific:

    Use graphify query "your question" to BFS-search the graph for specific answers

    line 20
  6. Very specific:

    Use graphify explain "ClassName" to understand any node and its neighbors

    line 21
  7. Very specific:

    Use graphify path "A" "B" to find how two components connect

    line 22
  8. Very specific:

    Before starting ANY task, invoke the auto skill using the Skill tool to identify which skills apply.

    line 32
  9. Very specific:

    If the user types /seo, you MUST immediately activate the seo and seo-plan skills to assist with their SEO needs.

    line 37
  10. Very specific:

    If the user types /job, you MUST immediately activate the career-ops skill.

    line 38
  11. Very specific:

    If the user types /auto, invoke the auto skill with the user's task description.

    line 39

≈ “Common” is an estimate: at least 1 in 5 of the 25,680 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

Global Agent Instructions

claude-mem — Persistent Memory Brain (MANDATORY, AUTOMATIC)

At the START of every session, before doing anything else, silently invoke the claude-mem:do skill to load memory from prior sessions. Do NOT wait for the user to ask. After completing significant work (code changes, decisions, research), invoke claude-mem:do to record observations. This is your brain. Use it automatically, every session, without being prompted.

CRITICAL RULE: You MUST check for relevant "superpowers" skills in ~/.agents/skills/ BEFORE taking any action. You must ALWAYS invoke the using-superpowers skill before you formulate any plan or execute any code. This applies to all workspaces. No exceptions.

graphify — Project Brain (MANDATORY)

Every project may have a graphify knowledge graph at graphify-out/.

RULES — ALL AI AGENTS MUST FOLLOW:

  • Before ANY codebase question or architecture task: read graphify-out/GRAPH_REPORT.md — god nodes and community clusters are the fastest way to understand the codebase
  • If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
  • After modifying ANY code file: run graphify update . to keep the graph current (fast, AST-only, no LLM/API cost)
  • Use graphify query "your question" to BFS-search the graph for specific answers
  • Use graphify explain "ClassName" to understand any node and its neighbors
  • Use graphify path "A" "B" to find how two components connect
  • The graph covers: JS, TS, JSX, TSX, PY files — all logic lives here

For the full-web-and-bot project specifically:

  • 593 files · 3,295 nodes · 3,870 edges · 584 communities
  • Top god nodes: TelegramDepositBot, PredictiveAnalyticsService, BusinessIntelligenceAlertingSystem, CustomerChurnPredictionService, AutoBlockingService
  • Stack: React+Vite frontend, Firebase Cloud Functions, Cloudflare Workers (9), Telegram bot, AI agents
  • Graph auto-updates on every git commit/checkout via installed hooks

Auto Skill Orchestration (MANDATORY)

Before starting ANY task, invoke the auto skill using the Skill tool to identify which skills apply. The auto skill will analyze the task, match it against all installed skills, and tell you which ones to chain together. Skip this only for one-word replies or pure conversation — any task involving code, design, research, deploy, review, or planning MUST go through auto first.

Custom Shortcuts

  • If the user types /seo, you MUST immediately activate the seo and seo-plan skills to assist with their SEO needs.
  • If the user types /job, you MUST immediately activate the career-ops skill.
  • If the user types /auto, invoke the auto skill with the user's task description.

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.