What is egregore AGENTS.md?

An agent protocol for Egregore, a workspace that supports several coding-agent runtimes, including Claude Code, Pi, Prime Agent, and Codex. It explains which instructions each runtime should follow and how agents communicate or hand off work.

Use it when starting work in Egregore, identifying the active agent harness, following the correct project instructions, or managing handoffs.

It prevents an agent from applying the wrong startup procedure or configuration for the runtime it is using.

How egregore-labs/egregore configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to egregore-labs/egregore. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/egregore-labs/egregore/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/egregore-labs/egregore

Made for: Claude Code, Codex, OpenCode.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/egregore-labs/egregore/agents-md.svg)](https://agentmods.dev/instructions/egregore-labs/egregore/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/egregore-labs/egregore/agents-md"><img src="https://agentmods.dev/badge/instructions/egregore-labs/egregore/agents-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 7,777 This file is loaded in full into every session.

When invoked 7,777 The same file — it is already loaded in full.

Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.07777	$0.07777
Opus 5	$0.03889	$0.03889
Sonnet 5	$0.01555	$0.01555
Haiku 4.5	$0.00778	$0.00778

Measured 3d ago against content hash ce95fcc291ed, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade B, and why

egregore AGENTS.md scanned grade B with 2 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.

Subtle steeringmediumPrompt injection

Instructions that bias recommendations or shape behaviour without the user noticing.

- Never tell the user to "ask their admin" for credentials. The user IS the admin.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Run `catalog` only when the route is unclear. Named reads return bounded stable fields and exact canonical `evidencePath` pointers; open only the returned files needed to answer. When a read reports partial coverage, use

AGENTS.md · 503 lines

How it starts

The opening of the file, as written. The whole thing — 503 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Egregore Agent Protocol

Runtime precedence — read this first.

If you are Claude Code (or any agent that loads .claude/): stop here. Your authoritative instructions are CLAUDE.md and .claude/. Follow the SessionStart greeting and the branch-on-first-message rule there. Do not follow the startup steps below and do not use bin/agent.sh for work an Egregore skill already covers. This file exists only for runtimes that cannot read CLAUDE.md.

If you are Pi (the pi coding-agent harness): stop here. Your authoritative Egregore instructions are loaded from .pi/APPEND_SYSTEM.md, and your workflows are exposed through the project-local .pi/ runtime. Do not follow the Codex-specific block below.

If you are Prime Agent (the prime-agent harness): stop here. Your authoritative Egregore instructions are loaded from .prime/agent/APPEND_SYSTEM.md, and your workflows are exposed through the project-local .prime/agent/ runtime. Do not follow the Codex-specific block below.

If you are Codex or another shell-only agent: this file is yours — continue.

Egregore is no longer only a Claude Code workspace. Claude Code remains the first-class integrated runtime through .claude/ and CLAUDE.md; any other agent that can run shell commands in this checkout participates through the portable memory protocol below.

Startup

Read egregore.json for the instance name, GitHub owner, and managed repos.
Run bin/agent.sh sync to pull the latest shared memory.
Read memory/people/ to learn collaborator handles.
Run bin/agent.sh activity --for <your-handle> to inspect handoffs and pending questions addressed to you.

Communication

Use the runtime-neutral bridge instead of Claude Code slash commands:

bin/agent.sh branch --topic "auth review"
bin/agent.sh save --message "Save: auth review" --topic "auth review" \
  --pr-body "$PR_BODY"   # body per .claude/context/pr-format.md (auto-skeleton if omitted)
bin/agent.sh wrap --from alice --topic "auth review" \
  --summary "Implemented the OAuth callback parser and documented follow-ups." \
  --body "Open threads: review error cases and browser redirects."

bin/agent.sh handoff --from alice --to bob --topic "auth review" \
  --body "Implemented the OAuth callback parser. Bob should review error cases."

bin/agent.sh ask --from bob --to alice --topic "auth review" \
  --question "Should invalid state redirect to login or return 400?"

bin/agent.sh answer --from alice \
  --question memory/knowledge/questions/2026-04-26-bob-to-alice-auth-review.md \
  --body "Return 400 in the API path; redirect only in browser routes."

Read the full file on GitHub · 503 lines

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.

3d ago Changed · +2 lines · +10 tokens per session ce95fcc291ed
7d ago First seen · 501 lines · 7,767 tokens per session scan B 1fecaf3925a7

Subscribe to this mod's changes

egregore AGENTS.md is an instructions file published in the GitHub repository egregore-labs/egregore (282 stars, last pushed 3d ago), licensed MIT. It adds 7,777 tokens to every session, about $0.0389 per session on Opus 5. A static security scan graded it B with 2 findings (subtle steering, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Claude Code Collaboration ai claude cli developer-tools

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

egregore: Instructions file for Claude Code