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/us1929/switchai/claude-mdgit clone --depth 1 https://github.com/US1929/switchaiWrote 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/us1929/switchai/claude-md)<a href="https://agentmods.dev/instructions/us1929/switchai/claude-md"><img src="https://agentmods.dev/badge/instructions/us1929/switchai/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 | $0.01779 | $0.01779 |
| Opus 5 | $0.00890 | $0.00890 |
| Sonnet 5 | $0.00356 | $0.00356 |
| Haiku 4.5 | $0.00178 | $0.00178 |
Grade A, and why
switchai 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 — 111 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.
SwitchAI — AI Agent per Cambio Fornitore Energia (Italia)
Panoramica
SwitchAI è un AI agent che automatizza il cambio fornitore Luce e Gas nel mercato libero italiano.
- Dominio: switchai.it (OVH Pro Web Hosting, PHP 8.5 + Apache, mod_rewrite)
- Stack: React 19 + Vite 8 + Tailwind CSS 4 (frontend) | PHP 8.5 API (backend) | Node.js MCP Server
- Design: ispirato a Switcho.it / Billoo.it, card allineate a ComparaSemplice
- Stile: inline
styleprops React (no CSS modules) - Dati: JSON flat-file con
flock()per locking concorrente, nessun database SQL - WebMCP: 4 tool registrati via
navigator.modelContext.registerTool()(Chrome 146+) - MCP Server: 6 tool per Claude Desktop + endpoint PHP a
/mcp
Comandi essenziali
# Sviluppo locale (servono ENTRAMBI i processi)
cd frontend && npm run dev # Vite dev server (porta 5173, proxy /api → 8080)
cd frontend/dist && php -S localhost:8080 router.php # PHP backend (in un altro terminale)
# Build per produzione
cd frontend && npm run build # Genera dist/ (frontend + PHP copiati automaticamente via copyBackendPlugin)
# Poi caricare TUTTO dist/ su OVH in www/
# Lint
cd frontend && npm run lint # ESLint
# MCP Server (Claude Desktop)
cd mcp-server && npm install && node index.js
Struttura chiave
frontend/dist/ ← BUILD OUTPUT da caricare su OVH (www/)
api/index.php ← API Router principale (include endpoint V2 /api/analyze)
mcp/index.php ← MCP Server PHP (JSON-RPC via HTTP POST)
inc/ ← Librerie PHP (tariff_loader, bill_parser, llm_logger, api_auth)
router.php ← Entry point PHP built-in server
data/ ← market_history.json, templates/ (JSON flat-file)
frontend/src/
pages/ ← Home, ComeFunziona, Analisi, Admin, Stats, Login, ConfermaRegistrazione, ...
components/ ← TariffCard, StickyReferenceBar, MarketSignal, ChatDemo, Navbar, Footer, ...
lib/
calc.js ← Calcoli tariffari (calcLuceCost, calcGasCost, getCurrentPricePerUnit, savingsToHuman, getRankingBadges, isPriceAnomalous, estimateRegulatedCosts)
api.js ← Chiamate API backend
webmcp.js ← Registrazione tool WebMCP (navigator.modelContext.registerTool)
validators.js ← Validazione form (codice fiscale, P.IVA, email, CAP)
province.js ← Mappatura provincia → zona (NORD/CENTRO/SUD)
App.jsx ← Router React (react-router-dom v7)
frontend/public/ ← File statici copiati in dist/ (llms.txt, webmcp.json, robots.txt, .htaccess, .well-known/, *.html)
backend/php/ ← Sorgenti PHP (specchiati in dist/ al build)
mcp-server/ ← MCP Server Node.js per Claude Desktop (index.js, server.json, Dockerfile)
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 · 111 lines · 1,779 tokens per session scan A 6df863db7006
switchai CLAUDE.md is an instructions file published in the GitHub repository US1929/switchai (0 stars, last pushed 13d ago), licensed MIT. It adds 1,779 tokens to every session, about $0.0089 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
ollama CLAUDE.md
Claude Code instructions for ollama/ollama: See AGENTS.md for the shared agent instructions for this repository.
llm AGENTS.md
Instructions for simonw/llm, covering agents.md, setting up a development environment and building the documentation.
plano CLAUDE.md
Instructions for katanemo/plano, covering claude.md, build & test commands, rust — wasm plugins (must target wasm32-wasip1), rust — brightstaff binary (native target) and rust — tests, format, lint.
oumi AGENTS.md
AGENTS.md instructions for oumi-ai/oumi, covering agent instructions, publication policy and working with contributions.
dev3000 AGENTS.md
Instructions for vercel-labs/dev3000, covering agents.md, runtime, local ui, production workflows, browser tools and development rules.
graphiti AGENTS.md
Instructions for getzep/graphiti, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.