What is open-work-relay AGENTS.md?

Work instructions for Agent Relay, a system that moves tasks through a visible queue so people and agents can continue them later. They explain how to classify work, record its source and state, and hand it off.

Use them when handling queued planning, production, content, support, scheduling, or code tasks that need a clear workflow and handoff.

They prevent important decisions, evidence, and task status from being kept only in a private conversation.

How The-Little-AI-Company/open-work-relay configures Codex and OpenCod

Reuse

Borrowing it

Nothing to install: this file belongs to The-Little-AI-Company/open-work-relay. 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/The-Little-AI-Company/open-work-relay/dev/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/The-Little-AI-Company/open-work-relay

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/the-little-ai-company/open-work-relay/agents-md/github.svg)](https://agentmods.dev/instructions/the-little-ai-company/open-work-relay/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/the-little-ai-company/open-work-relay/agents-md"><img src="https://agentmods.dev/badge/instructions/the-little-ai-company/open-work-relay/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/the-little-ai-company/open-work-relay/agents-md"><img src="https://agentmods.dev/badge/instructions/the-little-ai-company/open-work-relay/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 945 This file is loaded in full into every session.

When invoked 945 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	$0.00945	$0.00945
Opus 5.5	$0.00378	$0.00378
Sonnet 5.5	$0.00189	$0.00189
Haiku 4.5	$0.00094	$0.00094

Measured 2mo ago against content hash f50744095072, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

open-work-relay 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 · 181 lines

How it starts

The opening of the file, as written. The whole thing — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Relay Agent Protocol

You are operating inside a repo that uses Agent Relay.

Your job is not only to answer prompts. Your job is to move work through a visible queue with enough source, state, and receipt that another human or agent can continue without reading your private chat.

Core Distinction

Prompt mode asks for an answer.

Work mode gives a job to a system that can:

carry source material,
preserve decisions,
obey limits,
stop at gates,
leave evidence,
hand off cleanly.

When a task is in the relay, treat it as work mode.

Workflow Families

Before running meaningful work, classify the task into a workflow family.

Transcript-grounded families:

client call / follow-up,
support,
schedule / pickup change.

Agent Relay extension families:

planning,
production,
branding/content,
custom.

If the workflow is missing or unclear, propose one before acting.

Use:

templates/workflow.md
examples/workflows/

Do not force every job into the same process. A client follow-up, support ticket, and schedule change should follow the public transcript's shape closely. Planning, code edits, and brand review are Agent Relay extensions and should be proposed to the user as implementation choices.

Before You Start

Read:

The task record.
Any linked source material.
The queue adapter for the current queue.
The latest receipt, if one exists.

Do not begin if the task lacks:

outcome,
source material or a statement that no source exists,
allowed actions,
stop rules,
done evidence.

If any of those are missing, set the task to needs-input or leave a blocking question in the queue.

Claiming Work

Before editing or doing external work:

Confirm the task is eligible.
Set status to claimed or the queue-native equivalent.
Add your agent name and timestamp.
Leave a short claim receipt.
Move to working only when you actually start.

If another actor has already claimed the task, do not duplicate the work.

Read the full file on GitHub · 181 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 · 181 lines · 945 tokens per session scan A f50744095072

Subscribe to this mod's changes

open-work-relay AGENTS.md is an instructions file published in the GitHub repository The-Little-AI-Company/open-work-relay (4 stars, last pushed 4mo ago), licensed MIT. It adds 945 tokens to every session, about $0.0038 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.

Project management

open-work-relay AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
open-work-relay AGENTS.mdThe-Little-AI-Company/open-work-relay	Instructions file	4	4mo ago	MIT	A	945	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

open-work-relay: Instructions file for Codex