Local read-only Model Context Protocol server for Apache CouchDB
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"couchdb-mcp-server","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# CouchDB MCP Server CouchDB MCP Server is an open-source Python Model Context Protocol server for Apache CouchDB. The project is in early development. Version `0.1.0 Read-Only Preview` is implemented in source but has not been published as a release package yet. Published packages and release installation commands should not be assumed to exist until they are documented in a release. ## 0.1 Scope Version `0.1.0 Read-Only Preview` is a local, strictly read-only MCP server for Apache CouchDB 3.x. The approved version 0.1 constraints are: - local `stdio` MCP transport only; - one CouchDB instance per server process; - no writes, administration, replication, attachments, dashboard, embedded AI, or remote MCP server; - compact, bounded, structured, model-friendly responses; - security controls enforced by implementation, not only documentation. See [docs/SCOPE_0_1.md](docs/SCOPE_0_1.md) for the contributor-facing scope summary. ## 0.1 Tools Version 0.1 exposes exactly these nine product tools: 1. `get_server_status` 2. `list_databases` 3. `get_database_info` 4. `get_document` 5. `list_documents` 6. `query_documents` 7. `list_indexes` 8. `explain_query` 9. `inspect_database_structure` The `query_documents` implementation supports a safe Mango subset: `selector`, `fields`, `limit`, `bookmark`, `sort`, and `use_index`. It does not expose arbitrary Mango options. The `list_indexes` implementation lists Mango index metadata for authorized databases. It does not create, delete, or modify indexes. The `explain_query` implementation explains a controlled Mango query plan for an authorized database. It does not return matching documents or expose raw design documents. The `inspect_database_structure` implementation infers probable field structure from a bounded document sample. It reports field paths, observed types, occurrence counts, and frequencies without returning example values. All nine approved version 0.1 tools a…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.