MCP: Context7
by Upstash
Pulls up-to-date, version-specific documentation and code examples for libraries straight into your prompt.
Couchbase's documentation agent as a single MCP tool: ask a question about any Couchbase product, SDK or version and get an answer with source links. No cluster and no credentials.
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Couchbase Guru is a deliberately narrow MCP server from the Couchbase Ecosystem team. It exposes one tool, ask_couchbase_docs, which forwards a question to a hosted retrieval-augmented documentation agent and returns a natural-language answer followed by the documentation URLs it drew on.
Most database MCP servers connect an agent to your data. This one connects it to the manual. That distinction is the point: no Couchbase cluster, credentials or network access is required, because the server only talks to the documentation backend. An agent writing Couchbase code can consult current documentation without anyone provisioning it access to a database.
The backend keeps no conversation history, so questions have to be self-contained — include the product, version and SDK language, as in "How do I create a primary index with the Python SDK in Couchbase Server 7.6?" The answer arrives with source links, so the agent's claim is checkable against docs.couchbase.com rather than taken on trust.
Published on PyPI as couchbase-guru and runnable with uvx couchbase-guru — zero configuration, since the public documentation agent is the default backend. Python 3.10 or later plus uv. Source is on GitHub under the Apache 2.0 licence, and can be run from a clone with uv --directory if you would rather pin it.
Covered in the Weekly
by Upstash
Pulls up-to-date, version-specific documentation and code examples for libraries straight into your prompt.
by Prefect
Prefect's official read-only MCP server for inspecting flow runs, deployments and automations on Prefect Cloud or self-hosted, with a docs proxy for current CLI and SDK guidance.
by Fivetran
Fivetran's own MCP server exposes the Fivetran REST API to agents as discoverable tools, so an assistant can check sync status, inspect connections and destinations, and trigger syncs from chat.
Context-compression MCP server that shrinks tool outputs, logs, files and RAG chunks before they reach the model, cutting 60-95% of tokens on JSON payloads.