Borrowing it
Nothing to install: this file belongs to mcp-gtw/demo-game. 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/mcp-gtw/demo-game/main/CLAUDE.mdgit clone --depth 1 https://github.com/mcp-gtw/demo-gameWrote 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/mcp-gtw/demo-game/claude-md)<a href="https://agentmods.dev/instructions/mcp-gtw/demo-game/claude-md"><img src="https://agentmods.dev/badge/instructions/mcp-gtw/demo-game/claude-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/mcp-gtw/demo-game/claude-md"><img src="https://agentmods.dev/badge/instructions/mcp-gtw/demo-game/claude-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 | $0.10788 | $0.10788 |
| Opus 5 | $0.05394 | $0.05394 |
| Sonnet 5 | $0.02158 | $0.02158 |
| Haiku 4.5 | $0.01079 | $0.01079 |
Grade B, and why
demo-game CLAUDE.md scanned grade B with 1 finding 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 9d 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.
Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
(mcp.json, the pretty-printed JSON with one Copy button), Endpoint + token (the two shown separately, How it starts
The opening of the file, as written. The whole thing — 483 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for working in this repository.
What this is
mcp-gateway-demo is a multiplayer, top-down action game where every agent is controlled only
through MCP tools. It is the reference extension of
mcp-gtw: one AppGateway(Gateway) subclass adds an
authoritative grid world, its HTTP/WebSocket surface and a self-hosted Phaser client.
The internal package name is intentionally generic (app) so the demo is easy to port or reuse.
It depends on mcp-gtw. Locally [tool.uv.sources] points at the sibling ../mcp-server checkout.
In Docker/PyPI it resolves from the index.
Architecture (per-browser session, MCP-only login, camera follows your player)
The gateway is a relay. Each browser opens one session websocket (/app/stream) that gives it a
private per-session channel with an in-process provider (provider.py). The agent connects to
that channel's MCP endpoint and calls login, the provider adopts the resulting player for the
session, and the same websocket then streams the world. Everything after login is via MCP.
There is no browser login form — login is an MCP tool, so a player is created exactly once per
session, and there is no way to end up with two players. Positions are always grid cells (1, 2, 3…)
on the server (pixels exist only in the browser, tile_size is a render hint).
Flow: browser opens the socket with its stored token → server derives the stable channel and sends the
MCP url + token → agent connects and calls login → provider adopts the player, socket sends login +
catalog + map + snapshots → the page drops into the game following only that player (fixed zoom, no
picker, no zoom, no drag). Closing the browser tears the session down after a short grace, removing the
player (a reload within the grace keeps the same character; the token stays valid regardless). The
stats panel is fed by {type: "me"} over the same socket.
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.
- 9d ago First seen · 483 lines · 10,788 tokens per session scan B 9518835c1c97
demo-game CLAUDE.md is an instructions file published in the GitHub repository mcp-gtw/demo-game (3 stars, last pushed 1mo ago), licensed MIT. It adds 10,788 tokens to every session, about $0.0539 per session on Opus 5. A static security scan graded it B with 1 finding (reads mcp configuration). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.
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.