What is DarwinRelay AGENTS.md?

A set of repository instructions for coding agents working on DarwinRelay, a macOS runtime that lets software use tools such as shell commands, files, browsers, and desktop controls.

Use it as guidance before making non-trivial changes in the DarwinRelay repository.

It tells agents what the project does, which files to read first, and how to respect its authority and security boundaries.

How dcierra/DarwinRelay configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to dcierra/DarwinRelay. 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/dcierra/DarwinRelay/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/dcierra/DarwinRelay

Made for: 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/dcierra/darwinrelay/agents-md/github.svg)](https://agentmods.dev/instructions/dcierra/darwinrelay/agents-md)

Your own site

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

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/dcierra/darwinrelay/agents-md"><img src="https://agentmods.dev/badge/instructions/dcierra/darwinrelay/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,046 This file is loaded in full into every session.

When invoked 3,046 The same file — it is already loaded in full.

Security scan A 0 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	$0.03046	$0.03046
Opus 5.5	$0.01218	$0.01218
Sonnet 5.5	$0.00609	$0.00609
Haiku 4.5	$0.00305	$0.00305

Measured 2mo ago against content hash b53acd54a46a, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

DarwinRelay 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 2mo 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.

AGENTS.md · 229 lines

How it starts

The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

This file is the primary repository guide for coding agents working on DarwinRelay. Read it before modifying code. Human-facing product documentation starts in README.md; runtime architecture and security details live under docs/ and in SECURITY.md.

What DarwinRelay is

DarwinRelay is a macOS MCP execution runtime. It intentionally exposes broad local authority to an MCP client: shell/filesystem operations, real PTYs, long-running jobs, persisted Codex history, a managed background Chrome workspace, and native desktop control through Accessibility, ScreenCaptureKit, Vision and CoreGraphics.

It is not a sandbox. Security work in this repository is about making powerful authority explicit, bounded, auditable and fail-closed — not pretending arbitrary shell access is contained.

Read these files first

For any non-trivial task, inspect:

README.md — product surface and user workflows.
AGENTS.md — coding-agent rules and repository map.
SECURITY.md — trust boundaries and failure semantics.
docs/ARCHITECTURE.md — component/data-flow architecture.
docs/AGENT_OPERATIONS.md — runtime tool catalog and agent operating guidance.
docs/DESKTOP_CONTROL.md — native UI control model and limitations.
docs/DEVELOPMENT_MODEL.md — canonical public/private/upstream workflow and history mapping.
UPSTREAM.md — upstream lineage and attribution policy.

Repository map

bridge.mjs — MCP server, tool schemas, dispatch, shell/filesystem/jobs/Codex/native-UI integration and high-level policy.
mcp-http.mjs — authenticated HTTP/OAuth MCP transport.
lib/ — PTY/federation/browser/native helper clients and shared runtime modules.
desktop-helper/ — Swift native desktop-control helper and virtual cursor source.
chrome-extension/ — unpacked Chrome extension for the managed background browser workspace.
scripts/chrome-native-host.mjs — Native Messaging host implementation installed into Application Support.
menubar/ — Swift menu-bar app and bundle builder.
launchd/ — LaunchAgent templates.
scripts/ — install, signing, diagnostics, lifecycle, browser-profile and deployment helpers.
tests/ — deterministic protocol/integration/lifecycle tests plus the mutable native AppKit fixture.
.github/workflows/ci.yml — public GitHub-hosted CI split into descriptive jobs.

Read the full file on GitHub · 229 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.

2mo ago First seen · 229 lines · 3,046 tokens per session scan A b53acd54a46a

Subscribe to this mod's changes

DarwinRelay AGENTS.md is an instructions file published in the GitHub repository dcierra/DarwinRelay (4 stars, last pushed 2mo ago), licensed MIT. It adds 3,046 tokens to every session, about $0.0122 per session on Opus 5.5. 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.

accessibility ai-agents browser-automation chatgpt codex computer-use

DarwinRelay AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
DarwinRelay AGENTS.mddcierra/DarwinRelay	Instructions file	4	2mo ago	MIT	A	3,046	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

DarwinRelay: Instructions file for Codex