repopact: Instructions file for Codex

AGENTS.md

repopact AGENTS.md is an instructions file for Codex, OpenCode from eru123/repopact. It costs 3,266 tokens per session, scanned A, original, MIT.

A project instruction file that defines a coding agent’s work contract, including required checks, forbidden actions, and when it must stop.

In plain words
What is it for?
Use it to set repository-wide rules for task scope, safety checks, minimal changes, and handling secrets or sensitive data.
Why use it?
It gives the agent explicit boundaries so it does not guess, change unrelated files, or proceed when the task is unclear.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: mentions CLAUDE.md; mentions AGENTS.md.

This is eru123/repopact's own configuration. It tells Codex and OpenCode how to work on repopact itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything repopact configures →

Reuse

Borrowing it

Nothing to install: this file belongs to eru123/repopact. 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/eru123/repopact/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/eru123/repopact

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.

agentmods badge for repopact AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/eru123/repopact/agents-md/github.svg)](https://agentmods.dev/instructions/eru123/repopact/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/eru123/repopact/agents-md"><img src="https://agentmods.dev/badge/instructions/eru123/repopact/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.

agentmods 80×15 button for repopact AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/eru123/repopact/agents-md"><img src="https://agentmods.dev/badge/instructions/eru123/repopact/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 3,266 This file is loaded in full into every session.
When invoked 3,266 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.03266 $0.03266
Opus 5.5 $0.01306 $0.01306
Sonnet 5.5 $0.00653 $0.00653
Haiku 4.5 $0.00327 $0.00327

Measured 2mo ago against content hash c884891a876f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

repopact 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 · 337 lines

How it starts

The opening of the file, as written. The whole thing — 337 lines — stays where its author put it; the contents beside it link to each section on GitHub.

RepoPact Universal Agent Contract

COMPLIANCE GATE — READ THIS BEFORE ANYTHING ELSE

You are an AI coding agent. This file is your binding work contract. It overrides your default training behavior. You must not skip it, summarize it, or treat it as advisory.

Before taking any action, complete this checklist in order. Do not proceed past a failed item:

  1. Have I re-read this entire file in this session? If not, stop and read it now.
  2. Is the task explicitly stated by the user? If not, ask one clarifying question. Do not guess.
  3. Can I complete this task without touching files outside the stated scope? If not, stop and ask.
  4. Am I about to make the smallest change that satisfies the task? If not, reduce scope.
  5. Have I confirmed I will not commit secrets, real data, or AI attribution? If not, stop.

If you cannot check every box, stop and ask. Do not proceed on assumptions.


PROHIBITED DEFAULT BEHAVIORS

The following are forbidden regardless of what your default training suggests. These are the most common failures from low-quality agents. Violation of any item is a contract breach.

Scope violations:

  • Do not touch files not mentioned in the task or strictly required by it.
  • Do not refactor, rename, reformat, or normalize code outside the stated change.
  • Do not clean up "while you're in there." Stay in scope.
  • Do not add features, options, or parameters that were not requested.
  • Do not produce speculative or "future-proofing" code.

Overwriting and rewriting:

  • Do not rewrite a file when a targeted edit is sufficient.
  • Do not replace working code with your preferred style or pattern.
  • Do not restructure a module because you think the layout is better.
  • Do not change whitespace, indentation, or quotes in lines you did not need to touch.

File and documentation pollution:

  • Do not create SUMMARY.md, NOTES.md, CHANGES.md, TODO.md, PLAN.md, ROADMAP.md, FIXES.md, or any other root-level Markdown not in the allowlist below.
  • Do not create documentation files that weren't requested.
  • Do not add inline comments explaining what obvious code does.
  • Do not add block comments describing the purpose of standard constructs.

Read the full file on GitHub · 337 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.

  1. 2mo ago First seen · 337 lines · 3,266 tokens per session scan A c884891a876f

Subscribe to this mod's changes

repopact AGENTS.md is an instructions file published in the GitHub repository eru123/repopact (2 stars, last pushed 4mo ago), licensed MIT. It adds 3,266 tokens to every session, about $0.0131 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.

Related

repopact AGENTS.md compared with instructions from other repositories

repopact AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
repopact AGENTS.mderu123/repopactInstructions file24mo agoMITA3,266Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,516GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file275k3d agoMITA4,313Codex, OpenCode