acp-forge AGENTS.md

acp-forge AGENTS.md is an instructions file for Codex, OpenCode from ky2renzzz/acp-forge. It costs 626 tokens per session, scanned A, original, MIT.

A set of instructions for ACP Forge, a Rust workspace that helps coordinate coding agents, record and replay their work, manage Git worktrees, run skill pipelines, and verify web sources. A monorepo or workspace is one repository containing several related packages.

In plain words
What is it for?
Use it when working on ACP Forge’s six Rust crates, including orchestration, Git worktrees, execution graphs, skills, web verification, and its command-line tool.
Why use it?
It gives agents the project’s architecture, coding rules, and build and test expectations so their changes fit the repository.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/ky2renzzz/acp-forge/agents-md
Clone the repo
git clone --depth 1 https://github.com/ky2renzzz/acp-forge

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 acp-forge AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/ky2renzzz/acp-forge/agents-md.svg)](https://agentmods.dev/instructions/ky2renzzz/acp-forge/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/ky2renzzz/acp-forge/agents-md"><img src="https://agentmods.dev/badge/instructions/ky2renzzz/acp-forge/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 626 This file is loaded in full into every session.
When invoked 626 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.00626 $0.00626
Opus 5 $0.00313 $0.00313
Sonnet 5 $0.00125 $0.00125
Haiku 4.5 $0.00063 $0.00063

Measured 5d ago against content hash 420c1d757909, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

acp-forge 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 5d 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 · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ACP Forge — Agent Instructions

This file is read by Grok Build and compatible agents (Claude Code, Codex CLI). It defines how agents should work within this project.

Project Overview

ACP Forge is a Rust workspace that extends xAI Grok Build with:

  • Reproducible execution graphs (DAG recording + replay)
  • Pre-merge worktree divergence detection
  • Composable skill pipelines (TOML + SKILL.md)
  • Cross-reference web source verification

Architecture

Monorepo with 6 crates under crates/. Each crate is independently compilable.

Crate Purpose Key types
acp-forge-core Orchestration primitives Agent, Task, Orchestrator, Id<T>
acp-forge-worktree Git worktree management WorktreePool, WorktreeHandle
acp-forge-graph Execution DAG Session, Event, Recorder, Replayer
acp-forge-skills Skill templates Skill, Registry, Pipeline
acp-forge-web Web verification Provider, Cache, Verification
acp-forge-cli CLI binary clap subcommands

Conventions

  • Rust 2024 edition. Use edition = "2024" patterns.
  • Zero warnings policy. cargo check must produce no warnings.
  • Every public type gets a doc comment.
  • Tests next to code. Use #[cfg(test)] mod tests in each module.
  • Error types per crate. Each crate has its own error.rs with thiserror.
  • Typed IDs. Use Id<T> from acp-forge-core — never raw UUIDs.
  • No unwrap() in library code. Only in tests.
  • Imports: workspace dependencies via Cargo.toml, not path hacks.

Build & Test

cargo check          # Must pass with 0 warnings
cargo test           # Must pass all tests
cargo build --release  # For release binary

When Making Changes

  1. Run cargo check before committing — zero warnings required.
  2. Add tests for new public functions.
  3. Update the relevant crate's lib.rs re-exports if adding new modules.
  4. If adding a dependency, add it to workspace Cargo.toml first.
  5. Skill files: use TOML for parameterized skills, SKILL.md for simple prompts.

Read the full file on GitHub · 60 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. 5d ago First seen · 60 lines · 626 tokens per session scan A 420c1d757909

Subscribe to this mod's changes

acp-forge AGENTS.md is an instructions file published in the GitHub repository ky2renzzz/acp-forge (5 stars, last pushed 2mo ago), licensed MIT. It adds 626 tokens to every session, about $0.0031 per session on Opus 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.