Godot-MCP: Instructions file for Codex

AGENTS.md

Godot-MCP AGENTS.md is an instructions file for Codex, OpenCode from IvanMurzak/Godot-MCP. It costs 3,050 tokens per session, scanned B, original, Apache-2.0.

A set of project instructions for working on Godot-MCP, an add-on that lets AI coding agents communicate with the Godot game editor. It documents the code layout, build and test commands, and important project constraints.

In plain words
What is it for?
It is for building, testing, and maintaining a Godot game-development add-on that connects to an AI agent through the Model Context Protocol.
Why use it?
It gives an agent the repository-specific rules it needs to avoid changing shared dependencies, mixing editor and game code, or using the wrong checks.

Instructions file for CodexOpenCode

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

This is IvanMurzak/Godot-MCP's own configuration. It tells Codex and OpenCode how to work on Godot-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 Godot-MCP configures →

Reuse

Borrowing it

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

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/ivanmurzak/godot-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/ivanmurzak/godot-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 3,050 This file is loaded in full into every session.
When invoked 3,050 The same file — it is already loaded in full.
Security scan B 2 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 $0.03050 $0.03050
Opus 5.5 $0.01220 $0.01220
Sonnet 5.5 $0.00610 $0.00610
Haiku 4.5 $0.00305 $0.00305

Measured yesterday against content hash 802f7d334469, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

Godot-MCP AGENTS.md scanned grade B with 2 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 yesterday.

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.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

`--quit-after 600000`, then `curl -X POST <host>/api/system-tools/ping -d '{}'` → `pong`. `ping` is a

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

`--quit-after 600000`, then `curl -X POST <host>/api/system-tools/ping -d '{}'` → `pong`. `ping` is a
AGENTS.md · 138 lines

How it starts

The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — Godot-MCP

Godot-MCP connects LLM agents to the Godot editor over the Model Context Protocol — the Godot sibling of Unity-MCP. The MCP server is NOT in this repo: the addon consumes the shared GameDev-MCP-Server.

The addon is a [Tool] EditorPlugin (GodotMcpPlugin) that installs the main-thread dispatcher, builds a Reflector with Godot type converters, and connects to an MCP server (cloud ai-game.dev by default, or a custom local one) through the McpPlugin SignalR client. The editor tools' data layer is RQL (rql-query / rql-mutate / rql-schema, Editor/Rql/). The MCP/reflection/query stack (com.IvanMurzak.ReflectorNet, com.IvanMurzak.McpPlugin, com.IvanMurzak.Rql — PolyForm Shield 1.0.0) is not forked: it is consumed from nuget.org as PackageReferences whose pins are owned by the upstream release pipelines. Never bump them here. The current pins are in Godot-MCP.csproj (mirrored by Godot-MCP.Tests/, Godot-Tests/ and cli/src/utils/addon-deps.ts; keep them in lockstep) and restated in README.md / docs/.

Layout

  • addons/godot_mcp/ — the shipped addon (C#, Godot.NET.Sdk). Runtime/ is un-gated and compiles into game builds; Editor/ is the editor-only surface (#if TOOLS). Map and the Editor → Runtime dependency rule: docs/ARCHITECTURE.md. Do NOT put agent/dev files here: everything in it ships to consumers (release zip and Asset Library).
  • addons/godot_mcp/{Runtime,Editor}/Tools/ — runtime-safe / editor-driving tool families: one [AiToolType] partial class Tool_<Family> per family, one partial-class file per [AiTool("<name>")] + [Description] method. Names mirror Unity-MCP where sensible; full list in README "Tools Reference". New editor data access goes into RQL (an adapter, source, function or editor call under Editor/Rql/, see docs/rql.md), not a new tool.
  • Godot-MCP.Tests/ — xUnit CI suite; one <Subject>Tests.cs per area.
  • Godot-Tests/ — headless Godot CI testbed + runtime harness (Godot-Tests/README.md).
  • cli/ — the godot-cli npm package; see cli/AGENTS.md.
  • scripts/ — CI/release helpers (bump-version.mjs, check-runtime-boundary.py, harness asserts).
  • extensions/ — independent source-only NuGet modules; read its AGENTS.md before editing.
  • templates/extension/ — extension scaffolding; never included in core or consumer assemblies.
  • docs/ — ARCHITECTURE.md, RELEASING.md, runtime-security.md, Asset Library material.
  • glama/ + glama.json — the Glama directory listing image (stdio server + headless Godot editor in one container); not shipped, not CI. Admin-form settings in glama/README.md.
  • tests/chain-fixtures/, .github/scripts/chain_*.py — release-train fixtures; byte-compared, LF-pinned.

Read the full file on GitHub · 138 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. yesterday Changed · +2 lines · +57 tokens per session 802f7d334469
  2. 7d ago Changed · +2 tokens per session 0d1ac9222a38
  3. 8d ago Changed · +1 lines · +30 tokens per session abdeaec8cde2
  4. 9d ago Changed · +2 lines · +72 tokens per session dee7ce2aa5cc
  5. 10d ago Changed · +6 lines · +153 tokens per session 692da606c24c
  6. 17d ago Changed · +1 lines · +28 tokens per session 51d03254d1f1
  7. 18d ago Changed · +12 lines · +358 tokens per session 578774700954
  8. 20d ago Changed · +10 lines · +312 tokens per session 3c1004422a55
  9. 21d ago Changed · +8 lines · +210 tokens per session a4832c5bb7f0
  10. 25d ago First seen · 96 lines · 1,828 tokens per session scan B 58cb5e192831

Subscribe to this mod's changes

Godot-MCP AGENTS.md is an instructions file published in the GitHub repository IvanMurzak/Godot-MCP (325 stars, last pushed yesterday), licensed Apache-2.0. It adds 3,050 tokens to every session, about $0.0122 per session on Opus 5.5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-12.

Related

Godot-MCP AGENTS.md compared with instructions from other repositories

Godot-MCP AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
Godot-MCP AGENTS.mdIvanMurzak/Godot-MCPInstructions file325yesterdayApache-2.0B3,050Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode