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/manager/wolfs-basement/claude-mdgit clone --depth 1 https://github.com/manager/wolfs-basementWhat 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 | $0.04890 | $0.04890 |
| Opus 5 | $0.02445 | $0.02445 |
| Sonnet 5 | $0.00978 | $0.00978 |
| Haiku 4.5 | $0.00489 | $0.00489 |
Grade A, and why
wolfs-basement CLAUDE.md scanned grade A 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
The `wss.on('connection')` handler rejects any Origin that isn't `localhost`/`127.0.0.1` on the server's own `PORT` (see `isAllowedOrigin()` in server.js). Missing Origin is allowed (non-browser clients: node tests, curl How it starts
The opening of the file, as written. The whole thing — 182 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
Wolf's Basement is a visual RPG-themed terminal multiplexer that runs up to 4 Claude Code CLI sessions simultaneously. It has two modules: a pixel-art dungeon game (Basement) rendered in Phaser 3, and a multi-tab terminal interface (Terminal) — both served from a single Express + WebSocket server.
Commands
npm start # Start the server (default port 3000)
npm run dev # Same as npm start
node server.js # Direct launch
No build step. No bundler. No tests. The frontend is served as static files from public/.
Architecture
Three-file system
| File | Lines | Role |
|---|---|---|
server.js |
~1100 | Express server, WebSocket hub, Claude CLI process manager, git/auth/dev-server endpoints |
public/index.html |
~3500 | Terminal UI, agent cards, status bar, WebSocket client, all CSS inline |
public/game/main.js |
~3500 | Phaser 3 game scene — all rendering, animations, spatial awareness, agent behaviors |
Data flow
Browser (index.html + game/main.js)
↕ WebSocket (JSON messages)
Server (server.js)
↕ stdin/stdout (stream-json NDJSON)
Claude CLI processes (one per agent)
The server spawns Claude CLI with --output-format stream-json --input-format stream-json. It keeps the process alive between commands using --resume <sessionId>. User commands go in via proc.stdin.write(), responses stream back via proc.stdout as NDJSON lines.
CRITICAL — argument order in sendCommand: --resume <id> MUST be pushed BEFORE --allowed-tools. The CLI declares --allowed-tools <tools...> as variadic; if --resume <uuid> follows it, the parser may swallow --resume and the UUID as additional tool names, the resume silently fails, and the agent spawns a fresh session with no prior context. Symptom: after clicking "ALLOW ALL" on a permission prompt, the agent responds "Looking at the current state, it seems like you might have been in the middle of something..." Also pass tools as ONE comma-separated arg (tools.join(',')), not multiple positional args — that's the documented format and the safer pattern.
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.
- 2d ago First seen · 182 lines · 4,890 tokens per session scan A f1f4fce42747
wolfs-basement CLAUDE.md is an instructions file published in the GitHub repository manager/wolfs-basement (5 stars, last pushed 4mo ago), licensed MIT. It adds 4,890 tokens to every session, about $0.0244 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
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).
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.
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 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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.