ForgeMCP: Instructions file for Codex

AGENTS.md

ForgeMCP AGENTS.md is an instructions file for Codex, OpenCode from hardened-steel/ForgeMCP. It costs 2,080 tokens per session, scanned A, original, MIT.

A set of instructions for ForgeMCP, a Python server that gives coding agents controlled tools for C++ work such as building, testing, and debugging.

In plain words
What is it for?
It helps agents work on ForgeMCP while following its documentation, module boundaries, workspace limits, type-annotation rules, and test requirements.
Why use it?
It gives agents project rules for architecture, file access, logging, and development practices, reducing the risk of unsafe or inconsistent changes.

Instructions file for CodexOpenCode

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

This is hardened-steel/ForgeMCP's own configuration. It tells Codex and OpenCode how to work on ForgeMCP 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 ForgeMCP configures →

Reuse

Borrowing it

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

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 ForgeMCP AGENTS.md

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/hardened-steel/forgemcp/agents-md"><img src="https://agentmods.dev/badge/instructions/hardened-steel/forgemcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 2,080 This file is loaded in full into every session.
When invoked 2,080 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 $0.02080 $0.02080
Opus 5.5 $0.00832 $0.00832
Sonnet 5.5 $0.00416 $0.00416
Haiku 4.5 $0.00208 $0.00208

Measured 23d ago against content hash 1733534a82ac, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ForgeMCP 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 23d 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 · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ForgeMCP agent guide

Project

ForgeMCP is a Python MCP server for structured C and C++ development. The current codebase is a small foundation; do not recreate the deleted plugin/core framework.

Read first

Before designing or changing a module, read:

  1. README.md
  2. docs/architecture.md
  3. The nearest tests and the C++ acceptance fixture documentation when relevant

Design rules

  • Prefer direct code over framework-building. Add an abstraction only after at least two concrete callers need the same policy or lifecycle.
  • Do not add callbacks, interfaces, or other abstractions solely for tests. Keep operation-specific orchestration at its call site.
  • MCP-independent functions and structures are justified only by concrete inter-module interaction. Use the SDK directly for MCP-specific behavior.
  • src/forgemcp/server.py is the composition root. It creates services, injects their dependencies, and registers their MCP surface.
  • A service never constructs another service in its constructor. Dependencies are explicit constructor parameters and are wired in server.py.
  • New feature code goes in src/forgemcp/<feature>/. Put business logic and the main service class in service.py, expected domain errors in errors.py, and tests in tests/<feature>/test_service.py.
  • Extra files such as models.py, parsers, process runners, or kit discovery are allowed only when they represent a real responsibility that would otherwise make service.py harder to understand or test.
  • Keep public result models beside the service until their number or reuse justifies models.py.
  • Use WorkspacePath for paths shared between modules and in tool arguments/results. It serializes as project/..., storage/..., or root/<absolute-path>; native executable locations and low-level filesystem operations still use Path.
  • Each feature module owns filesystem operations on its files. Do not read, write, create, move, or delete files through WorkspaceService methods from another module. Implement Workspace tool operations in their decorated MCP handlers; expose only the Workspace APIs other modules actually need for provider registration and result resources.
  • Feature services launch external programs only through the injected ProcessService; do not call asyncio.create_subprocess_exec directly outside the process module.

Read the full file on GitHub · 174 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. 23d ago Changed · +109 lines · +1,614 tokens per session 1733534a82ac
  2. 1mo ago Changed · +15 lines · +113 tokens per session a80c68b9825f
  3. 1mo ago First seen · 50 lines · 353 tokens per session scan A 7e20c5117898

Subscribe to this mod's changes

ForgeMCP AGENTS.md is an instructions file published in the GitHub repository hardened-steel/ForgeMCP (0 stars, last pushed 7d ago), licensed MIT. It adds 2,080 tokens to every session, about $0.0083 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

Other instructions, from other repositories