mongo-conventions

What problem does it solve?

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.

Core Features & Use Cases

• Async Motor Access Patterns: Enforce non-blocking async driver usage for all request-path database operations, with centralized connection setup and idempotent index creation.
• Append-Only Data Lake Discipline: Protect irreplaceable raw source data by treating downstream derived collections as replayable transforms, eliminating data loss from bulk reprocessing.
• Mandatory Provenance Tracking: Ensure all stored data includes source and timestamp metadata, making aggregated results traceable to primary sources and avoiding incorrect default values.
• Use Case: When building a content ingestion pipeline that stores crawled web data in MongoDB, this skill ensures raw payloads are never lost, derived embeddings can be rebuilt for free if deleted, and every content chunk is traceable to its original source URL and fetch timestamp.

Quick Start

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.