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.
npx agentmods add instructions/beremaran/godot-agent-loop/claude-mdgit clone --depth 1 https://github.com/beremaran/godot-agent-loopWrote 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/beremaran/godot-agent-loop/claude-md)<a href="https://agentmods.dev/instructions/beremaran/godot-agent-loop/claude-md"><img src="https://agentmods.dev/badge/instructions/beremaran/godot-agent-loop/claude-md.svg" alt="Measured on agentmods" 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 | $0.01380 | $0.01380 |
| Opus 5 | $0.00690 | $0.00690 |
| Sonnet 5 | $0.00276 | $0.00276 |
| Haiku 4.5 | $0.00138 | $0.00138 |
Grade A, and why
godot-agent-loop CLAUDE.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.
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.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What this is
Godot Agent Loop (@beremaran/godot-agent-loop): a TypeScript MCP server (stdio transport) that lets agents author, run, observe, and verify Godot 4 games. Entry point src/index.ts, compiled to build/index.js. Godot 4.7 is both the compatibility floor and primary target.
Commands
npm run build # tsc + metadata/adapter sync + scripts/build.js (copies GDScript into build/)
npm test # unit tests (vitest)
npx vitest run tests/tool-registry.test.ts # single unit test file
npx vitest run tests/tool-registry.test.ts -t "name" # single test by name
npm run test:e2e # retained full-path MCP E2E (builds first, needs Godot)
npm run build && npx vitest run --config vitest.e2e.config.ts tests/e2e/representative-path.test.ts # single E2E file
npm run lint # eslint + markdownlint; npm run lint:md:fix auto-fixes markdown
npm run check # lint + build + retained unit suite; used by CI and release validation
npm run check:built # lint + retained unit suite, assuming a fresh build
npm run check:fast # lint + tsc --noEmit — what the Husky pre-commit hook runs (~seconds)
npm run inspector # MCP Inspector against build/index.js
Godot binary resolution for E2E tests: GODOT_BIN (executable path) → godot4/godot on PATH → first godot* executable inside GODOT_PATH (a directory). Note: the MCP server itself treats GODOT_PATH as an executable path, not a directory — the E2E harness passes the binary explicitly.
Architecture
Tool pipeline (TypeScript side)
Every MCP tool flows through this chain:
src/tool-definitions.ts— name + JSON schema for all tools.src/tool-manifest.ts— one entry per tool: domain, handler method, backend, action list, privilege flag. Typed asRecord<ToolName, ...>so completeness is a compile-time fact.src/domain-tool-registries.ts— dispatches to the three handler classes insrc/tool-handlers/: lifecycle (processes/editor), project (validation, imports, integrity, .NET, add-ons, exports), game (runtime interaction).src/tool-surface.ts— the advertised surface. Default is a lean 20-toolcoreset covering the author → validate → run → observe → interact → assert loop;GODOT_MCP_TOOL_SURFACE=fulladvertises the complete 64-tool catalog. Thegodot_catalogtool searches/describes the full catalog andgodot_calldispatches a hidden tool by name either way. Size budgets (bytes/token/count) are constants here.
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.
- 5d ago First seen · 60 lines · 1,380 tokens per session scan A 01b270cb784a
godot-agent-loop CLAUDE.md is an instructions file published in the GitHub repository beremaran/godot-agent-loop (5 stars, last pushed yesterday), licensed MIT. It adds 1,380 tokens to every session, about $0.0069 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.
Other instructions, from other repositories
claudemon CLAUDE.md
Instructions for zamarrowski/claudemon, covering ai agent guidelines, project shape, general guidelines, general rules and guards & defensive code.
unity-code-style-guide AGENTS.md
Instructions for krogh-jacobsen/unity-code-style-guide, covering agents.md — unity 6 c, project setup — edit this block, never do these — they corrupt the project, deprecated in unity 6 and if you read nothing else.
nes-php-glfw copilot-instructions.md
Instructions for oliverearl/nes-php-glfw, covering github copilot development guidelines for nes emulator, agent document source of truth, project overview, code style & standards and documentation.
game-and-watch-retro-go-sd CLAUDE.md
Claude Code instructions for sylverb/game-and-watch-retro-go-sd, covering claude.md, what this project is, build / flash workflow, architecture and three storage tiers, one elf.
base-building-trap-defense-design-agent-skill CLAUDE.md
Instructions for dungnotnull/base-building-trap-defense-design-agent-skill, covering claude.md — skill 250: base-building-trap-defense-design, skill identity, problem this skill solves, harness flow summary and sub-skills.
trident-mcp AGENTS.md
Instructions for mordor-forge/trident-mcp, covering agents.md, what this is, build and test, single-file verification and e2e tests.