openagents: Instructions file for Codex

AGENTS.md

openagents AGENTS.md is an instructions file for Codex, OpenCode from OpenAgentsInc/openagents. It costs 18,186 tokens per session, scanned A, original, Apache-2.0.

Repository instructions for AI agents working on the OpenAgents Rust CLI and related services. They define which parts of the repository are current, which are historical, and which language and documentation rules apply.

In plain words
What is it for?
Working on the Rust CLI, preserving the transcript archive, avoiding the historical web tree, and writing covered public documentation using the project's approved terminology.
Why use it?
They prevent agents from changing retired web code or adding TypeScript to a Rust-only repository. They also set a simplified-English standard for specified public documentation.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: reads .claude/ paths; mentions subagents; mentions Claude Code.

This is OpenAgentsInc/openagents's own configuration. It tells Codex and OpenCode how to work on openagents 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 openagents configures →

Reuse

Borrowing it

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

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 openagents AGENTS.md

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/openagentsinc/openagents/agents-md"><img src="https://agentmods.dev/badge/instructions/openagentsinc/openagents/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 18,186 This file is loaded in full into every session.
When invoked 18,186 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 $0.18186 $0.18186
Opus 5.5 $0.07274 $0.07274
Sonnet 5.5 $0.03637 $0.03637
Haiku 4.5 $0.01819 $0.01819

Measured today against content hash 7ea28aeb8597, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

openagents 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 today.

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 · 1,100 lines

How it starts

The opening of the file, as written. The whole thing — 1,100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OpenAgents agent contract

Product code in this workspace is Rust. Do not add TypeScript. The existing browser exception is apps/haunt-preview: the owner explicitly requested Effect, TypeScript, and direct WebGPU for the browser-first Haunted House shader/Kiln demo (#12500, 2026-10-22) and its explicitly requested Medieval Town Lab. Keep this exception scoped to those creator previews; reuse Kiln and existing domain services rather than duplicating them. The owner also requested the /new-lander marketing preview and its Effect/direct-WebGPU universe selector (2026-10-22). Marketing and routes stay in Rust; apps/verse-selector is the scoped TypeScript browser island. It reuses the existing account and world services and does not own domain logic. The owner explicitly authorized React, Effect Atom, Effect, and shadcn/ui in apps/verse-ui for the complete Verse presentation (2026-10-23, #12576). Keep account, project, publish, invitation, and engine authority in the existing Rust services and controllers; the UI does not receive private engine source. The existing product exception is swift/lev-bridge, the helper that reaches Apple's FoundationModels framework, which has no Rust binding; it is built by a repo script and supervised as a child process. Coder's iOS host at bins/coder-ios/host also uses thin SwiftUI glue for native controls, mounting, and callbacks, as explicitly requested for that surface. Keep its application state, domain logic, permissions, and transport in Rust; the implemented observer keeps these in coder-mobile and coder-connect. Read crates/rust-native/docs/spec.md and docs/coder/rust-native/architecture.md before adding that boundary. The OpenAgents desktop and mobile apps at apps/openagents/ (GPUI, UIKit, and Kotlin shells over Rust) are the named exception described in "One OpenAgents desktop app and one mobile app" below. The Android host at bins/coder-android/host uses the equivalent thin Kotlin boundary for Android framework widgets, SurfaceView, camera, sensors, and Keystore access. Keep domain state, Nostr, authorization, cache, and world behavior in the same Rust mobile library; do not import the private Android backend or authentication implementation. Retained Python training and acceptance tooling and shell orchestration are infrastructure exceptions, not permission to add another product implementation language. The Nix and shell under os/ (CoderOS) are infrastructure in the same sense: they configure a machine and launch Rust programs, and product behavior belongs in Rust.

The current OpenAgents mobile app lives in apps/openagents. Read apps/openagents/OPENAGENTS.md for builds and apps/openagents/apps/ios/release/README.md for TestFlight releases. The previous mobile hosts and their dedicated Rust crate were removed in #11901.

Read the full file on GitHub · 1,100 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. today Changed · +4 lines · +72 tokens per session 7ea28aeb8597
  2. yesterday Changed · +8 lines · +149 tokens per session e9e0260a873b
  3. 2d ago Changed · +9 lines · +163 tokens per session 63b6e799ab57
  4. 3d ago Changed · +5 lines · +113 tokens per session 7111e9599a9f
  5. 4d ago Changed · +11 lines · +228 tokens per session 48983bb054d9
  6. 5d ago Changed · +54 lines · +850 tokens per session 23126fea3b0f
  7. 6d ago Changed · +26 lines · +440 tokens per session d5b49a1a4301
  8. 7d ago Changed · +37 lines · +651 tokens per session 8355da7aaa35
  9. 8d ago Changed · +17 lines · +241 tokens per session 02dad1abe7cf
  10. 9d ago First seen · 929 lines · 15,279 tokens per session scan A 484ab86559ba

Subscribe to this mod's changes

openagents AGENTS.md is an instructions file published in the GitHub repository OpenAgentsInc/openagents (458 stars, last pushed today), licensed Apache-2.0. It adds 18,186 tokens to every session, about $0.0727 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 2026-08-30.

Related

Other instructions, from other repositories

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 · 9,715 tokens

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 · 7,581 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).

microsoft/vscode · 5,262 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,680 tokens

deepseek-harness AGENTS.md

AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.

deepseek-ai/deepseek-harness · 4,261 tokens

vscode copilot-instructions.md

Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).

microsoft/vscode · 2,731 tokens