Turns natural language into MongoDB find queries or aggregation pipelines by first pulling collection schemas, indexes, and sample documents through the MongoDB MCP server. It validates field names against the schema before generating anything, which saves you from the silent failures MongoDB loves to give when you typo a field. The skill defaults to simple find queries over aggregations when possible and includes practical guardrails like avoiding redundant operators and filtering early in pipelines. Worth noting it only handles read operations and won't touch Atlas Search, vector queries, or query optimization, so scope is deliberately narrow here.
npx skills add https://github.com/mongodb/agent-skills --skill mongodb-natural-language-querying