Godot AI is a Godot add-on and Model Context Protocol server that connects AI assistants to a live Godot editor. It is used to let coding agents create and modify game scenes, scripts, nodes, and other engine settings through the editor.
Borrowing it
Nothing to install: this file belongs to hi-godot/godot-ai. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/hi-godot/godot-ai/main/AGENTS.mdgit clone --depth 1 https://github.com/hi-godot/godot-aiWrote 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.
[](https://agentmods.dev/instructions/hi-godot/godot-ai/agents-md)<a href="https://agentmods.dev/instructions/hi-godot/godot-ai/agents-md"><img src="https://agentmods.dev/badge/instructions/hi-godot/godot-ai/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.
<a href="https://agentmods.dev/instructions/hi-godot/godot-ai/agents-md"><img src="https://agentmods.dev/badge/instructions/hi-godot/godot-ai/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.07615 | $0.07615 |
| Opus 5.5 | $0.03046 | $0.03046 |
| Sonnet 5.5 | $0.01523 | $0.01523 |
| Haiku 4.5 | $0.00762 | $0.00762 |
Grade A, and why
godot-ai 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 26d 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.
How it starts
The opening of the file, as written. The whole thing — 338 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Godot AI
This guide is for any AI assistant working in this repository. Keep Claude-specific files such as .claude/CLAUDE.md and .claude/skills/* as thin pointers to this shared guidance.
PR activity monitoring
Whenever a PR is created, opened, or identified in the conversation, check its state and start monitoring without asking again, unless it is already closed or merged. This authorizes read-only monitoring and notifications, not posting comments, changing code, rerunning CI, or merging.
- Discover the current environment's tools. Prefer a native PR activity
subscription (such as
subscribe_pr_activity) when actually available; confirm success and retain its subscription identifier. Do not assume a particular tool namespace or that GitHub access includes subscriptions. - If subscriptions are unavailable but durable task automations are supported,
reuse an existing monitor for the same repository and PR or create a
15-minute heartbeat attached to the current task. Poll PR comments, inline
review comments, submitted reviews, CI results, merge conflicts, and PR state
using the authenticated GitHub connector or
gh. Keep the last seen state, report existing actionable feedback once, and notify only on new actionable feedback, failures, conflicts, access failures, or closure/merge. Stay quiet otherwise; stop the monitor after closure/merge. - If neither capability is available, check current feedback and CI, then explicitly report that persistent monitoring was not started and which capability is missing. A one-time read, shell polling process, GitHub email subscription, or written instruction is not a durable task subscription.
- Report the confirmed monitoring mechanism when starting it. Never claim monitoring is active without a successful subscription/automation response.
This policy applies to local and cloud sessions. Check capabilities separately in each environment: local Codex settings and credentials do not establish cloud tool availability. Do not add a guessed MCP endpoint or install an integration merely because it has GitHub in its name. A future provider must be verified to expose the subscription and deliver events to the task.
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.
- 26d ago Changed · +42 tokens per session d5033b506f48
- 1mo ago Changed 2da1ae071a9c
- 1mo ago Changed · +16 lines · +219 tokens per session babe3774e31b
- 1mo ago Changed · +3 lines · +112 tokens per session d7cfca971e9b
- 1mo ago Changed · +32 lines · +393 tokens per session 8e06c8bd7ad2
- 1mo ago Changed · +4 lines · +54 tokens per session ddf52dc64154
- 1mo ago Changed · +73 tokens per session a31e323e44a9
- 1mo ago Changed · +24 lines · -440 tokens per session 64410713755b
- 2mo ago First seen · 259 lines · 7,162 tokens per session scan A da7292baabec
godot-ai AGENTS.md is an instructions file published in the GitHub repository hi-godot/godot-ai (3,264 stars, last pushed yesterday), licensed MIT. It adds 7,615 tokens to every session, about $0.0305 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.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.