Skip to content
Kitbay

mongodb-search-and-ai

Apache-2.0Updated on Nov 18, 2026206GitHub starsIncludes scripts

MongoDB's skill for search features on MongoDB Atlas: full-text Atlas Search, semantic Vector Search and hybrid search that combines the two.

What the skill teaches

MongoDB's skill for search features on MongoDB Atlas: full-text Atlas Search, semantic Vector Search and hybrid search that combines the two. The agent inspects databases, collections, schemas and indexes through the MongoDB MCP Server, works out which search type fits the use case, and then designs the index and query, including autocomplete, fuzzy matching, facets, filters, embeddings and retrieval for RAG. It can also run a guided tour on Atlas sample data. The tour needs an Atlas cloud cluster; Atlas Local and self-managed servers are out of its scope. The skill folder includes a Python script.

Capabilities

  • Choose between Atlas Search, Vector Search and hybrid search
  • Create search and vector search indexes
  • Write full-text, autocomplete and fuzzy search queries
  • Set up semantic search and RAG retrieval with embeddings
  • Run a guided search walkthrough on Atlas sample data