What is acp-cli AGENTS.md?

Project instructions for acp-cli, a Rust command-line program that starts coding agents and communicates with them through structured messages. ACP means Agent Client Protocol.

Developing and maintaining the acp-cli commands, agent connections, configuration, sessions, queues, permissions, and output formats.

They explain the project’s design, files, and coding rules before changes are made. This helps an agent work within the existing command, session, and queue structure.

How motosan-dev/acp-cli configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to motosan-dev/acp-cli. 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/motosan-dev/acp-cli/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/motosan-dev/acp-cli

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/motosan-dev/acp-cli/agents-md/github.svg)](https://agentmods.dev/instructions/motosan-dev/acp-cli/agents-md)

Your own site

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

Per session 1,147 This file is loaded in full into every session.

When invoked 1,147 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	$0.01147	$0.01147
Opus 5.5	$0.00459	$0.00459
Sonnet 5.5	$0.00229	$0.00229
Haiku 4.5	$0.00115	$0.00115

Measured 2mo ago against content hash ed4c88d50361, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

acp-cli 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 · 134 lines

How it starts

The opening of the file, as written. The whole thing — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — acp-cli Development Brief

Read this before writing any code.

Project Overview

acp-cli is a headless CLI client for the Agent Client Protocol (ACP). It spawns coding agents (Claude, Codex, Gemini, etc.) as child processes and communicates over structured JSON-RPC via stdio pipes.

Version: 0.2.1
Language: Rust (edition 2024)
Published to: crates.io (acp-cli)
GitHub: https://github.com/motosan-dev/acp-cli

Repository Structure

acp-cli/
├── Cargo.toml
├── src/
│   ├── main.rs            # CLI entry point
│   ├── lib.rs             # Library re-exports
│   ├── config.rs          # Config loading + merging
│   ├── error.rs           # Error types (thiserror)
│   ├── agent/registry.rs  # Agent name → command mapping
│   ├── bridge/            # AcpBridge (spawn_blocking + LocalSet)
│   ├── cli/               # Commands: init, prompt, exec, sessions
│   ├── client/            # BridgedAcpClient + permission resolution
│   ├── output/            # Renderers: text, json, quiet
│   ├── queue/             # Queue owner/client, IPC, lease
│   └── session/           # Session scoping, persistence, history
├── skills/acp-cli/        # AI skill card
├── README.md
├── llms.txt               # LLM-consumable API reference
├── CLAUDE.md              # Claude Code project instructions
├── CHANGELOG.md
└── LICENSE

Key Design Decisions

!Send bridge pattern — ACP SDK uses Rc/spawn_local; bridge runs in spawn_blocking + LocalSet, communicates via mpsc channels
Queue system — first process becomes queue owner (holds agent connection), subsequent processes connect as clients via Unix socket
Session scoping — key is SHA-256(agent + "\0" + git_root + "\0" + name), ensuring isolation per project/agent/name
Auth token chain — env var → config file → ~/.claude.json → macOS Keychain (never guess, always explicit)
Permission modes — three modes (approve-all, approve-reads, deny-all) applied at BridgedAcpClient level
Agent registry — hardcoded defaults + config overrides + raw command fallback

Read the full file on GitHub · 134 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 · 134 lines · 1,147 tokens per session scan A ed4c88d50361

Subscribe to this mod's changes

acp-cli AGENTS.md is an instructions file published in the GitHub repository motosan-dev/acp-cli (14 stars, last pushed 7mo ago), licensed MIT. It adds 1,147 tokens to every session, about $0.0046 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.

Languages Rust

acp-cli AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
acp-cli AGENTS.mdmotosan-dev/acp-cli	Instructions file	14	7mo ago	MIT	A	1,147	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

acp-cli: Instructions file for Codex