Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/ihabkhaled/clawai/agents-mdgit clone --depth 1 https://github.com/ihabkhaled/ClawAIWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.02353 | $0.02353 |
| Opus 5 | $0.01177 | $0.01177 |
| Sonnet 5 | $0.00471 | $0.00471 |
| Haiku 4.5 | $0.00235 | $0.00235 |
Grade A, and why
ClawAI AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ClawAI — Agent Entrypoint (canonical router)
This is the compact router every AI coding agent reads first. It does not
copy policy — it points at the canonical sources. If a tool-specific file
(CODEX.md, cursor.md, KIMI.md, …) disagrees with this file or the sources
below, the canonical source wins.
Repository identity
ClawAI — a local-first AI orchestration platform. 18 NestJS microservices + a
Next.js 16 frontend + 6 shared packages, in an npm-workspace monorepo. Events
flow over RabbitMQ (claw.events); nginx reverse-proxies /api/v1/*.
Canonical authority (higher wins on conflict)
CLAUDE.md— long-form operating policyrules/00-non-negotiable-rules.md— engineering blockerscontext/architecture-map.md— structural architecturecontext/stack-and-toolchain.md— commands + toolchain- Numbered
rules/*→skills/*→context/*+memory/* - Generated
.ai/manifests/*— machine-readable facts - This file and the other compact routers
First command (before touching code)
npm run knowledge:context -- --task="<what you are about to do>"
Then read .ai/local/current-context.md — it names the affected
workspaces, governing rules, matching skills, reviewers, related events/env/
permissions, files to inspect, the validation lane, and known pitfalls.
Full compact bootstrap: .ai/BOOTSTRAP.md (generated, ~800 tokens).
Where things live
- Architecture →
context/architecture-map.md· Commands →context/stack-and-toolchain.md - Service catalog →
context/service-catalog.md(data:.ai/manifests/services.json) - Rules →
rules/· Skills →skills/· Reviewer roles →agents/ - Durable pitfalls/lessons →
memory/· Task packs →.ai/packs/ - SDLC templates →
docs/features/_template/· ADRs →docs/13-adr/
Validation (touched folders only — never all 24 workspaces)
npm run affected:list # what your diff touches + why
cd <workspace> && npm run typecheck && npm run lint && npm test && npm run build
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 191 lines · 2,353 tokens per session scan A 1dda71b3de77
ClawAI AGENTS.md is an instructions file published in the GitHub repository ihabkhaled/ClawAI (22 stars, last pushed 3d ago), licensed Apache-2.0. It adds 2,353 tokens to every session, about $0.0118 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
Khala-Agentic-AI-Teams CLAUDE.md
Instructions for brandonkindred/Khala-Agentic-AI-Teams, covering claude.md, project overview, repository structure, architecture and build & run.
MC-GPT AGENTS.md
AGENTS.md instructions for IoTAiTech/MC-GPT: Author: Dr.-Ing. Babak Sorkhpour, with AI assistance Version: 6.8.0-beta.1.
ai-management-toolkit AGENTS.md
AGENTS.md instructions for wei-wei-hu/ai-management-toolkit: Use the repo-scoped skill at .agents/skills/ai-management-toolkit/SKILL.md for weekly team planning, meeting preparation, blocker diagnosis, feedback and coaching, SWOT/5W2H/root-cause/PDCA decisions, performance reviews, one-on-ones, SMART goals, WBS task…
guardian-agent CLAUDE.md
Instructions for Threat-Vector-Security/guardian-agent, covering repository guidelines, branching (critical), project structure & module organization, build, test, and development commands and intent gateway (critical).
ai-methods CLAUDE.md
Claude Code instructions for pirona/ai-methods, covering claude.md — claude code operator template, 1. identity & operator context, 2. homelab infrastructure, hypervisor & compute and network.
gameforge AGENTS.md
Instructions for robcost/gameforge, covering general guidelines for working with nx, scaffolding & generators and when to use nxdocs.