agentmods.dev

cortex CLAUDE.md

A Claude Code plugin and open protocol for coordinating AI agents through shared Markdown files. It defines team roles, onboarding, registration, and offboarding.

In plain words
What is it for?
It helps set up agent teams, distinguish coordinators from workers, and adapt the protocol for tools such as Codex, Cursor, Gemini CLI, and OpenCode.
Why use it?
It gives agents a shared place and agreed process for communicating without requiring a separate coordination service.

Instructions file

12 4mo ago A scan 463 tokens original MIT
View the source file agentweave/cortex
Install

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.

agentmods
npx agentmods add instructions/agentweave/cortex/claude-md
Clone the repo
git clone --depth 1 https://github.com/agentweave/cortex
Per session 463 This file is loaded in full into every session.
When invoked 463 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00463 $0.00463
Opus 5 $0.00231 $0.00231
Sonnet 5 $0.00093 $0.00093
Haiku 4.5 $0.00046 $0.00046

Measured 2d ago against content hash c64596e7955f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cortex CLAUDE.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.

CLAUDE.md · 41 lines

What it actually says

Cortex

An open protocol and Claude Code plugin for coordinating AI agent teams through plain markdown files.

Project Structure

skills/                  — Claude Code skills (the plugin)
  setup/          — first-time setup
  join/           — agent onboarding (worker + coordinator .cortex.md)
  leave/          — agent offboarding
  register/        — agent registration
docs/
  protocol.md            — standalone protocol spec (runtime-agnostic)
  adapters/              — onboarding examples for Codex, Cursor, Gemini CLI, OpenCode
  internal/              — specs and plans (gitignored)
.claude-plugin/
  plugin.json            — plugin metadata
  marketplace.json       — marketplace definition for installation

Key Concepts

  • Team directory — a shared folder of markdown files where agents coordinate. Not part of this repo — created by /setup.
  • Skills are prompts, not code — each SKILL.md is instructions that Claude Code follows. No runtime, no build step.
  • Protocol vs plugin — docs/protocol.md is the universal spec. The skills are one implementation of it.
  • Coordinator vs Worker .cortex.md — /join generates different protocols based on whether the agent slug is chief-of-staff or not. Both templates live in skills/join/SKILL.md.

Conventions

  • Commit messages: feat:, fix:, docs:, chore: prefixes
  • Skills are self-contained — each SKILL.md includes all templates inline rather than referencing external files
  • Config lives at ~/.cortex/config.yaml, not in the repo
  • .cortex.md and CLAUDE.local.md are gitignored — they're generated per-agent, not part of the plugin

When Editing Skills

  • Skill files are markdown with YAML frontmatter (name, description, user_invocable, argument-hint)
  • Steps are numbered and sequential — renumber if you insert or remove steps
  • Templates (agent notes, .cortex.md) are embedded inline with {placeholder} syntax
  • After editing, reinstall the plugin to pick up changes: claude plugins install cortex@agentweave --scope user
Changes

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.

  1. 2d ago First seen · 41 lines · 463 tokens per session scan A c64596e7955f

Subscribe to this mod's changes

cortex CLAUDE.md is an instructions file published in the GitHub repository agentweave/cortex (12 stars, last pushed 4mo ago), licensed MIT. It adds 463 tokens to every session, about $0.0023 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.

Building agentsagent-coordinationai-agentsclaude-codeclaude-code-plugincoordinationmarkdown
Related

Other instructions, from other repositories

skill-based-architecture AGENTS.md

Instructions for WoJiSama/skill-based-architecture, covering agents.md, quick routing (survives context truncation), auto-triggers and red flags — stop.

WoJiSama/skill-based-architecture · 829 tokens

research-os AGENTS.md

Instructions for lxinfei5/research-os, covering researchos — constitution, §0 what this is, §1 directory discipline, §2 half-life (innovation 1) — memory design and §3 behavior pillars (innovations 2–5).

lxinfei5/research-os · 1,143 tokens

markview CLAUDE.md

Instructions for paulhkang94/markview, covering markview — ai development guide, quick reference, architecture, mcp tools (for ai integration) and safety rules.

paulhkang94/markview · 1,604 tokens

codex copilot-instructions.md

Instructions for bocan/codex, covering codex - ai coding instructions, architecture overview, mcp server architecture, critical commands and docker deployment.

bocan/codex · 2,462 tokens

grit AGENTS.md

Instructions for rtk-ai/grit, covering persistent memory (icm), recall (before starting work), store (after completing significant work) and other commands.

rtk-ai/grit · 242 tokens

Cecil AGENTS.md

AGENTS.md instructions for Cecilapp/Cecil, covering cecil - agent instructions, start here, critical commands, architecture at a glance and coding conventions.

Cecilapp/Cecil · 718 tokens