Enforce MongoDB implementation patterns for async Python services.
npx skills add https://github.com/eric-sabe/engsys --skill mongo-conventions
Please help me install this Agent Skill. Skill: mongo-conventions Source: https://github.com/eric-sabe/engsys/tree/main/stacks/db/mongo/skills/mongo-conventions Command: npx skills add https://github.com/eric-sabe/engsys --skill mongo-conventions
Inconsistent MongoDB implementation patterns lead to data loss, event loop blocking in async services, unrecoverable derived data, and untraceable data provenance in projects using MongoDB.
Use the mongo-conventions skill to implement async Motor database access, set up an append-only raw data lake collection, and add provenance fields to all new MongoDB documents for your project's ingest pipeline.
High-intent search queries and answers about installing and using this skill.
Prevent event loop blocking in async Python MongoDB services by enforcing non-blocking Motor async driver usage for all request-path database operations, centralized connection setup, and idempotent index creation to maintain async concurrency.
Append-only data lake discipline in MongoDB protects irreplaceable raw source data by treating downstream derived collections as replayable transforms, eliminating data loss risks during bulk reprocessing of ingest pipelines.
Add provenance metadata to MongoDB documents by ensuring all stored data includes mandatory source and timestamp metadata, making aggregated results traceable to primary sources and avoiding incorrect default values.
Yes, the Motor async driver supports idempotent index creation for MongoDB by enforcing centralized connection setup and non-blocking async access patterns, ensuring indexes are created safely without blocking the event loop.
The best way to structure a MongoDB data lake for crawled web data is to enforce append-only raw payload collections, ensuring derived embeddings can be rebuilt for free and every content chunk remains traceable to its original source URL and fetch timestamp.
Inconsistent MongoDB indexing and implementation patterns cause unrecoverable derived data because without append-only raw lake discipline and provenance tracking, bulk reprocessing can overwrite or lose the original source payloads needed to replay transforms.
FindMalekConfigure Prisma for multiple database providers and set up the client.
Yangel-hideStore embeddings and metadata in an open-source vector database for semantic search.
levanminhducDesign MongoDB schemas, CRUD operations, and aggregation pipelines.
interserverReview MongoDB schemas for anti-patterns and embedding-versus-referencing decisions.
