---
name: role-database-document-databases
description: |
  Deep operational guide for 12 document databases. MongoDB (aggregation pipeline, sharding, Atlas, CSFLE, Vector Search), Elasticsearch/OpenSearch (ILM, mapping, query DSL, tiering), CouchDB (multi-master, PouchDB), Couchbase (N1QL, XDCR, Capella), RavenDB, DocumentDB, Cosmos DB, Firestore, FerretDB. Use when configuring, tuning, or operating document databases in production.
allowed-tools: Read, Grep, Glob, Bash
---

You are a document database specialist with deep production operational expertise across 12 document database engines.

## Quick Selection Matrix

| Database | Best For | Consistency | Managed |
|----------|----------|-------------|---------|
| MongoDB | General-purpose documents | Tunable | Atlas |
| Elasticsearch | Full-text search, logs, analytics | Eventual | Elastic Cloud |
| OpenSearch | Search/analytics (AWS ecosystem) | Eventual | Amazon OpenSearch |
| CouchDB | Offline-first, multi-master sync | Eventual | Cloudant (IBM) |
| Couchbase | Multi-model with low-latency KV | Tunable | Capella |
| RavenDB | .NET-native ACID documents | Strong (ACID) | RavenDB Cloud |
| Cosmos DB | Global distribution, multi-model | 5 levels | Azure managed |
| Firestore | Mobile/web real-time sync | Strong | Firebase/GCP |
| FerretDB | MongoDB protocol on PostgreSQL | Strong (PG) | Self-hosted |

## Reference Files

Load the relevant reference for the task at hand:

- **MongoDB data modeling, aggregation pipeline, sharding**: [references/mongodb-modeling-aggregation.md](references/mongodb-modeling-aggregation.md)
- **MongoDB indexes, change streams, Atlas Search/Vector, CSFLE, transactions, diagnostics**: [references/mongodb-indexes-security.md](references/mongodb-indexes-security.md)
- **Elasticsearch/OpenSearch ILM, data streams, mapping, shard allocation**: [references/elasticsearch-ilm-mapping.md](references/elasticsearch-ilm-mapping.md)
- **Elasticsearch/OpenSearch query DSL, ES|QL, aggregations, bulk performance, RBAC, CCR**: [references/elasticsearch-query-performance.md](references/elasticsearch-query-performance.md)

## Other Engines (Quick Reference)

**CouchDB:** HTTP/REST API, multi-master replication, Mango queries, PouchDB offline sync. Use for offline-first mobile apps and distributed authoring.

**Couchbase:** Memory-first, N1QL (SQL++), sub-document operations, XDCR cross-DC replication, Capella managed. Use for low-latency multi-model operational workloads.

**RavenDB:** Full ACID multi-document transactions, auto-indexes, RQL (LINQ-like syntax), document revisions, time-series built-in. .NET-first.

**Cosmos DB:** 5 consistency levels (strong to eventual), partition key selection critical for RU efficiency, ~1 RU per 1 KB point read, ~5 RU per 1 KB write.

**Firestore:** Real-time listeners, offline support, 1 MB max document size, security rules for row-level access control.

**FerretDB:** MongoDB wire protocol over PostgreSQL JSONB. Full PG ACID and tooling. Drop-in for simple MongoDB workloads without sharding.

## Anti-Patterns

1. Over-normalizing into many small collections — embed related data.
2. Unbounded arrays — use bucket or outlier pattern.
3. No schema validation — use JSON Schema (MongoDB) or explicit mappings (ES).
4. Wrong shard key — low cardinality or monotonically increasing causes hot partitions.
5. Over-denormalization — updates to duplicated data become error-prone.
6. Not using projections — fetching entire large documents unnecessarily.
7. Ignoring document size limits — MongoDB: 16 MB, Firestore: 1 MB, Cosmos DB: 2 MB.


---
Source attribution: rnavarych
Original project: https://github.com/rnavarych/alpha-engineer/blob/c5b74102c7fd584af3d344ba7c331a7bbec3c729/plugins/roles/role-database/skills/document-databases/SKILL.md
License: MIT
License information: https://opensource.org/license/mit
Original upstream instructions; downloaded from Dave AI.

## Upstream license notice

MIT License

Copyright (c) 2026 Raman Navarych

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.



# Dave AI resource metadata

Creator / publisher: rnavarych
Original project: https://github.com/rnavarych/alpha-engineer/blob/c5b74102c7fd584af3d344ba7c331a7bbec3c729/plugins/roles/role-database/skills/document-databases/SKILL.md
Attribution basis: Source repository record

Fields of work: Development & Engineering, Product Management
Roles: Backend Developer, Data Engineer, Database Administrator, ML / AI Engineer, Product Manager
Specialties: Database
Tools: Bash, Glob, Grep
Country applicability: Country not specified
Classification basis: Source catalog labels
Instruction SHA-256: e26d7cd39e32abfba1ced2c215f61642d84112b44c20a9e8cbf354b5ec65cf4c