Couchbase MCP Server
Couchbase MCP Server is a Model Context Protocol (MCP) server implementation that lets LLMs directly interact with data stored in Couchbase clusters through a rich set of tools.
It exposes capabilities like cluster health checks, schema discovery, document key-value operations, SQL++ querying, and query performance analysis in a controlled way that is safe by default via read-only modes and fine-grained tool disabling. The server supports multiple transport modes, including stdio and Streamable HTTP, so it can be run locally or shared across teams.
Architecture
Tools
The server exposes several tools across multiple categories.
See the Tools page for full details.
| Category | Tools |
|---|---|
| Cluster Setup & Health | get_server_configuration_status, test_cluster_connection, get_cluster_health_and_services |
| Data Model & Schema Discovery | get_buckets_in_cluster, get_scopes_in_bucket, get_collections_in_scope, get_scopes_and_collections_in_bucket, get_schema_for_collection |
| Document KV Operations | get_document_by_id, upsert_document_by_id, insert_document_by_id, replace_document_by_id, delete_document_by_id |
| Query and Indexing | run_sql_plus_plus_query, explain_sql_plus_plus_query, list_indexes, get_index_advisor_recommendations |
| Query Performance Analysis | get_longest_running_queries, get_most_frequent_queries, get_queries_not_selective, get_queries_not_using_covering_index, get_queries_using_primary_index, get_queries_with_largest_response_sizes, get_queries_with_large_result_count |
Tutorials
Build an AI agent using LangChain and the Couchbase MCP Server.AI Agent with OpenAI Agents SDK
Build an AI agent using the OpenAI Agents SDK and the Couchbase MCP Server.
Releases
The latest release is available on PyPI and Docker Hub.
See the Release Notes for version history and details.
The latest version (v0.7.0) introduces support for explaining queries and elicitation for tool calls.
Support Policy
We truly appreciate your interest in this project! This project is Couchbase community-maintained, which means it's not officially supported by our support team. However, our engineers are actively monitoring and maintaining this repo and will try to resolve issues on a best-effort basis.
Our support portal is unable to assist with requests related to this project, so we kindly ask that all inquiries stay within GitHub.
- Bug reports: Open a GitHub issue
- Feature requests: Open a GitHub issue with the "enhancement" label
- Questions: Open a GitHub issue
Your collaboration helps us all move forward together - thank you! Pull requests and contributions from the community are welcome and encouraged. See the contribution guidelines for more details: