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/gemini-mdgit clone --depth 1 https://github.com/ihabkhaled/ClawAIWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/ihabkhaled/clawai/gemini-md)<a href="https://agentmods.dev/instructions/ihabkhaled/clawai/gemini-md"><img src="https://agentmods.dev/badge/instructions/ihabkhaled/clawai/gemini-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.01379 | $0.01379 |
| Opus 5 | $0.00690 | $0.00690 |
| Sonnet 5 | $0.00276 | $0.00276 |
| Haiku 4.5 | $0.00138 | $0.00138 |
Grade A, and why
ClawAI GEMINI.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 4d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ClawAI — Gemini router
Compact router for Gemini. This file does not duplicate policy — it routes you to canonical sources. Canonical wins on conflict (see below).
Identity
ClawAI: 18 NestJS services + Next.js 16 frontend + 6 shared packages (npm
workspaces). RabbitMQ claw.events; nginx proxies /api/v1/*.
Canonical authority (higher wins)
CLAUDE.md2.rules/00-non-negotiable-rules.md3.context/architecture-map.mdcontext/stack-and-toolchain.md5. numberedrules/*→skills/*→context/*+memory/*- generated
.ai/manifests/*7. compact routers (this file)
First command (always)
npm run knowledge:context -- --task="<your task>"
Then read .ai/local/current-context.md. Compact bootstrap: .ai/BOOTSTRAP.md.
Validation (touched folders only)
npm run affected:list
cd <workspace> && npm run typecheck && npm run lint && npm test && npm run build
Release: npm run release:preflight.
Absolute prohibitions
- NEVER bypass git hooks (
--no-verify). NEVER suppress lint/types (eslint-disable,@ts-ignore,any,as unknown as). - NEVER cross a service DB boundary (use HTTP/RabbitMQ); no logic in controllers;
no DB calls outside repositories; no
process.envoutside AppConfig; noconsole.log. - NEVER log/expose secrets. NEVER add text without i18n (13 locales). NEVER add code without a test.
- Do NOT invent repository facts — derive them from
.ai/manifests/and real code. - NEVER ship a change with no knowledge delta — docs/skills/rules/context ship in the SAME commit (rules/33).
- NEVER gate per-commit or all-workspace; gate once at the end, scoped. Never re-prove an unchanged tree (rules/34).
Gemini emphasis
- Preserve canonical authority and repository-grounded facts. When you state a
port/event/route, cite the manifest it came from (
.ai/manifests/*). Do not paraphrase policy from memory — link the canonical rule.
See AGENTS.md for the full router. Detail lives in the canonical sources.
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.
- 4d ago First seen · 112 lines · 1,379 tokens per session scan A c42fa0de8e3e
ClawAI GEMINI.md is an instructions file published in the GitHub repository ihabkhaled/ClawAI (22 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,379 tokens to every session, about $0.0069 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
Claude Code instructions for brandonkindred/Khala-Agentic-AI-Teams, covering claude.md, project overview, repository structure, architecture and build & run.
herald CLAUDE.md
Claude Code instructions for kOlapsis/herald, covering claude.md — règles du projet herald, 🔒 règles absolues (non négociables), 📦 le projet, 🏗️ architecture and principes fondamentaux.
DQIII8 CLAUDE.md
Claude Code instructions for senda-labs/DQIII8, covering dqiii8 — architecture kernel, routing tiers (cost-first — strict), system map, rule engine and inviolable rules.
multi-agent-ralph-loop CLAUDE.md
Claude Code instructions for alfredolopez80/multi-agent-ralph-loop, covering multi-agent ralph v3.0.0, analysis methodology, configuration location, browser automation (v3.0) and security hooks.
continuum CLAUDE.md
Claude Code instructions for shyftlabs/continuum, covering claude code configuration - ruflo v3, behavioral rules (always enforced), file organization, project architecture and project config.
ai-agents push-lock.instructions.md
Instructions for rjmurillo/ai-agents, covering one push lock path, must, must not, historical measurements and commit guard.