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/darshjoshi/pitwall/claude-mdgit clone --depth 1 https://github.com/darshjoshi/pitwallWhat 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 | $0.01019 | $0.01019 |
| Opus 5 | $0.00509 | $0.00509 |
| Sonnet 5 | $0.00204 | $0.00204 |
| Haiku 4.5 | $0.00102 | $0.00102 |
Grade A, and why
pitwall 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 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.
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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — Pitwall Project
What This Is
Pitwall is an F1 MCP server with 79 tools. It provides Formula 1 data to Claude via the Model Context Protocol.
Key Architecture
pitwall.py— MCP server (79 tools). Auto-degrades: 30 tools without FastF1 (incl. 16 no-auth live-timing tools), 79 with it. Live car telemetry + GPS are auth-gated and full-only.signalr_client.py— Raw SignalR Core WebSocket client for live race data during sessions.decompressor.py— Zlib decompression for CarData.z (telemetry) and Position.z (GPS).merger.py— Deep-merge for F1's incremental JSON delta format.topics.py— All 20 SignalR topics with auth/compression metadata.auth_setup.py— F1 TV Premium token setup (browser OAuth flow).
Data Sources (all free, no API keys)
- F1 Static API (
livetiming.formula1.com/static/) — 33 feeds per session, 2018-present - Jolpica-F1 (
api.jolpi.ca/ergast/f1/) — Historical results 1950-present - FastF1 (optional) — Enhanced telemetry, plots, deep analysis
- SignalR Core (live) — Real-time during races, requires F1 TV auth for telemetry
F1 TV Authentication
Token stored at ~/.f1token and ~/Library/Application Support/fastf1/f1auth.json.
Expires every ~4 days. Re-authenticate: python3 auth_setup.py.
Only needed for live telemetry (CarData.z, Position.z) during active sessions.
All post-session and historical data is free without auth.
Sprint Weekend Gotcha
At sprint weekends, both Sprint and Race have Type="Race" in F1's data. The session resolver uses 3-pass priority (exact name > partial name > type) to correctly resolve "Race" to the main Race, not the Sprint. If you hit issues, pass the explicit session_path from list_races.
Dependency Pin — Do Not Loosen
mcp is pinned >=1.0.0,<2 in pyproject.toml and both requirements files. MCP Python SDK 2.0
removed mcp.server.fastmcp, which every one of the 79 tools is built on. Dropping the <2
ships a package that dies on import for every new user (this happened in 1.1.0). Lifting the pin
means porting all 79 tools to mcp.server.mcpserver.MCPServer first.
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 · 78 lines · 1,019 tokens per session scan A b65797e47059
pitwall CLAUDE.md is an instructions file published in the GitHub repository darshjoshi/pitwall (8 stars, last pushed 1mo ago), licensed MIT. It adds 1,019 tokens to every session, about $0.0051 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
racehooks-mcp AGENTS.md
AGENTS.md instructions for racehooks/racehooks-mcp, covering agents.md — racehooks mcp server, install (end user), what it exposes, security model and contributing to this repo.
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).