Mongodb

This gives Claude the mental model for MongoDB schema design and query patterns, which matters because the embedded vs referenced decision shapes everything else. The skill covers the core tradeoffs: when to denormalize data into a single document versus splitting it across collections with references. You get the aggregation pipeline for analytics, indexing strategies for performance, and practical examples of CRUD operations with all the query operators. The progressive disclosure flags when MongoDB makes sense (flexible schemas, hierarchical data) and when it doesn't (complex joins, strict ACID requirements). If you're building anything that stores JSON-like documents or needs to scale horizontally, this gives Claude the context to help you avoid the common pitfalls around document size limits and unbounded arrays.