
This connects Claude to a pre-deployment LLM cost calculator that models retries, prompt caching, batch discounts, and infrastructure overhead across 18 models from OpenAI, Anthropic, Google, DeepSeek, and others. Three tools: estimate_cost runs a full forecast with naive, realistic, and worst-case monthly spend; get_model_prices pulls current per-million-token rates with cache and batch pricing; list_archetypes returns seven preset patterns like RAG pipeline or multi-step agent. Works keyless in demo mode with six free models. Unlock the full catalog with a Pro key from beforeyouship.dev. Use it when you're designing a new LLM feature and need honest cost projections before writing code or picking a model tier.