forge-mcp: Instructions file for Codex

AGENTS.md

forge-mcp AGENTS.md is an instructions file for Codex, OpenCode from fuwadog/forge-mcp. It costs 1,725 tokens per session, scanned A, original, MIT.

Development instructions for forge-mcp, an MCP server project designed to run on Bun on Windows. MCP is a standard way for AI agents to connect to tools, but this project currently contains design documents and reference files rather than runnable source code.

In plain words
What is it for?
Planning and implementing forge-mcp in the required milestone order. It also helps prevent runtime state and bearer tokens from being stored in the repository.
Why use it?
It tells an agent which files, runtime, build constraints, and secret-storage rules to follow while implementing the project.

Instructions file for CodexOpenCode

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

This is fuwadog/forge-mcp's own configuration. It tells Codex and OpenCode how to work on forge-mcp 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 forge-mcp configures →

Reuse

Borrowing it

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

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 forge-mcp AGENTS.md

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/fuwadog/forge-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/fuwadog/forge-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,725 This file is loaded in full into every session.
When invoked 1,725 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 $0.01725 $0.01725
Opus 5.5 $0.00690 $0.00690
Sonnet 5.5 $0.00345 $0.00345
Haiku 4.5 $0.00172 $0.00172

Measured 2mo ago against content hash d3154b49bb57, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

forge-mcp 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 · 98 lines

How it starts

The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.

forge-mcp — Agent Guide

What this repo is

forge-mcp is a pre-implementation MCP server project. Only design docs and reference files exist until milestones land — no runnable source yet.

  • README.md — architecture overview, tool surface, guarantees
  • implementation_plan.md — the source of truth: LLM-executable build plan (M0–M7), gates, rollbacks
  • reference_files/ — nav-mcp source to port (READ-ONLY; never edit)
  • forge.example.json — config template (live forge.json is gitignored)

Target runtime & locations

  • Language: TypeScript (strict) · Runtime: Bun ≥ 1.3.x — not Node · OS: Windows 11, PowerShell 7
  • Code root: this repo. All source is committed here.
  • Runtime state: %LOCALAPPDATA%\forge-mcp\ — daemon.json (contains the bearer token), boot.lock, logs/. Never in the repo. If daemon.json ever appears under the repo root, that is a bug (and a secret leak) — stop and fix.

Key constraints (from implementation_plan.md §0)

Constraint Detail
bun build --compile BANNED — run TS source directly (SDK subpath-export issues)
Dependencies EXACT pins only. No ^/~. bun.lock committed. Manual bumps. Re-run M0 spike after any Bun bump.
Never-throw Every tool returns a NavEnvelope (ok:false on error; transport never sees exceptions)
reference_files/ Read-only port source. Live nav-mcp install untouched until M7 cutover.
opencode.json NEVER modified by agent. Snippets delivered for manual user apply.
Security floor 127.0.0.1 bind · bearer token · Origin validation/403 · .env* guard · canonical+realpath path jail · cmd-shim metachar gate
Invalidation authority Fingerprints (mtime-in-key) = correctness. Watcher events = eviction accelerator only.
Fallback floor Worst case == nav-mcp parity (embedded mode, stat invalidation, CLI/ripgrep ast). Below that = build failure.

Build order (sequential, gated — do not skip)

Read the full file on GitHub · 98 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 · 98 lines · 1,725 tokens per session scan A d3154b49bb57

Subscribe to this mod's changes

forge-mcp AGENTS.md is an instructions file published in the GitHub repository fuwadog/forge-mcp (0 stars, last pushed 5mo ago), licensed MIT. It adds 1,725 tokens to every session, about $0.0069 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

forge-mcp AGENTS.md compared with instructions from other repositories

forge-mcp AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
forge-mcp AGENTS.mdfuwadog/forge-mcpInstructions file05mo agoMITA1,725Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,516GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA2,022GitHub Copilot