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/agenticfirst/cerebro/agents-mdgit clone --depth 1 https://github.com/AgenticFirst/CerebroWhat 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 | $0.01287 | $0.01287 |
| Opus 5 | $0.00643 | $0.00643 |
| Sonnet 5 | $0.00257 | $0.00257 |
| Haiku 4.5 | $0.00129 | $0.00129 |
Grade A, and why
Cerebro 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 2d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cerebro — agent context
Cerebro is an Electron + React + TypeScript desktop app with a Python/FastAPI backend that drives AI workflows from natural-language chat. Inference runs through the Claude Code CLI as a subprocess; routines, experts, and integrations are persisted in SQLite under the user's app.getPath('userData').
Read before touching integrations
docs/adding-integrations.md is the single source of truth for wiring a new connected service (Slack, Gmail, Notion, GitHub, …). Adding an integration touches ~15 files across the bridge, IPC, UI, engine actions, routine prompt, and chat skills. Skipping any of them silently breaks part of the surface — chat-driven setup works, but the integration never appears in routines, or the chat agent can't tool-call it, etc. Work the playbook top-to-bottom.
The currently-implemented integrations — Telegram, HubSpot, WhatsApp — are the reference examples. Copy from whichever auth model is closest.
Project layout
| Path | Purpose |
|---|---|
src/main.ts |
Electron main process. Spawns the Python backend, owns IPC handlers, lifecycle. |
src/preload.ts |
Exposes the typed window.cerebro.* API to the renderer. |
src/types/ipc.ts |
IPC channel constants + API interfaces — the contract between main and renderer. |
src/components/ |
React UI. Chat lives in chat/, settings/integrations in screens/. |
src/agents/ |
Agent runtime that wraps Claude Code subprocesses. |
src/claude-code/installer.ts |
Writes the Cerebro main agent prompt and skill markdown into <userData>/.claude/. Where most chat-skill prompt logic lives. |
src/engine/ |
Routine DAG executor + actions + dry-run stubs. |
src/integrations/ |
Central registry for connected services (manifests + IPC wrappers). |
src/<service>/ |
Per-service bridges/holders (telegram/, hubspot/, whatsapp/). |
src/chat-actions/ |
Loopback HTTP server the chat subprocess calls back into. |
backend/ |
FastAPI service. SQLite, conversations, routines, memory, experts. |
docs/tech-designs/ |
Architecture references (routines.md, cerebro-core-intelligence.md, …). |
docs/adding-integrations.md |
Integration playbook — read this first. |
docs/roadmap.md |
Implementation status. |
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.
- 2d ago First seen · 78 lines · 1,287 tokens per session scan A 62b0f5ce9dc7
Cerebro AGENTS.md is an instructions file published in the GitHub repository AgenticFirst/Cerebro (8 stars, last pushed 1mo ago), licensed MIT. It adds 1,287 tokens to every session, about $0.0064 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 2026-08-31.
Other instructions, from other repositories
nextclaw AGENTS.md
Instructions for Peiiii/nextclaw, covering nextclaw ai 常驻内核, 产品愿景, 沟通与推进, 深思与目标模式 and 协作与 git 安全.
hope-agent AGENTS.md
AGENTS.md instructions for shiwenwen/hope-agent, covering hope agent, 安全红线, 提交前检查（强制）, 分支与发布 and 设置约定.
ovo AGENTS.md
Instructions for dushaobindoudou/ovo, covering agents.md, project overview, common commands, development and build.
Abu-Cowork AGENTS.md
AGENTS.md instructions for PM-Shawn/Abu-Cowork, covering abu (阿布) — ai desktop office assistant, project overview, tech stack, git workflow & development constraints and branches.
P-ai AGENTS.md
AGENTS.md instructions for kawayiYokami/P-ai, covering agents.md, 项目概述, 构建与开发命令, 开发模式（前端热重载 + rust 自动重编译） and 仅启动前端 dev server（端口 1420）.
Abu-Cowork CLAUDE.md
Claude Code instructions for PM-Shawn/Abu-Cowork, a project described as: Open-source alternative to Claude Cowork — a local-first AI agent desktop app · multi-model · self-evolving skills · privacy-first · multi-Harness roadmap · DeepSeek Harness integration in progress.