Mongodb MCP Server vs Ravendb MCP
A side-by-side look at the Mongodb MCP Server and Ravendb MCP MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Mongodb MCP Server for databases work as a local process; choose Ravendb MCP for databases as a local process. You can also run both side by side in one mcpServers block.
Which MCP server should you choose?
Choose Mongodb MCP Server when
- You need dedicated capabilities in the Databases domain.
- You prefer a local stdio subprocess transport.
- You have the keys it needs:
MDB_MCP_CONNECTION_STRING. - You want tools like
aggregate, aggregate-db, collection-indexes.
Choose Ravendb MCP when
- You need dedicated capabilities in the Databases domain.
- You prefer a local stdio subprocess transport.
- You have the keys it needs:
RAVENDB_URLS. - You want tools like
list_compare_exchange, get_database_stats, collect_debug_package.
Feature & specification comparison
Rows where the two servers differ are marked.
| Spec | ||
|---|---|---|
| Category | Databases | Databases |
| Quality signal | 65/100 · Great | 60/100 · Good |
| Install command | Confirmed | Confirmed |
| Transport | Local subprocess (stdio) | Local subprocess (stdio) |
| Auth | Not declared | Not declared |
| Pricing | Not declared | Not declared |
| Required env vars | MDB_MCP_CONNECTION_STRING | RAVENDB_URLS |
| Verified owner | Community listing | Community listing |
| GitHub stars | 1,146 | 2 |
| Directory activity | 1 views 0 installs 0 upvotes | 1 views 0 installs 0 upvotes |
| Tools | 53 | 21 |
| Full listing | View Mongodb MCP Server | View Ravendb MCP |
Tools & capabilities
Mongodb MCP Server tools 53
aggregateRun an aggregation against a MongoDB collectionaggregate-dbRun an aggregation against a MongoDB databasecollection-indexesDescribe the indexes for a collectioncollection-schemaDescribe the schema for a collectioncollection-storage-sizeGets the size of the collectionconnectConnect to a MongoDB instancecountGets the number of documents in a MongoDB collection using db.collection.count() and query as an optional filter parametercreate-collectionCreates a new collection in a database. If the database doesn't exist, it will be created automatically.create-indexCreate an index for a collectiondb-statsReturns statistics that reflect the use state of a single databasedelete-manyRemoves all documents that match the filter from a MongoDB collectiondisconnectClose a MongoDB connection and revoke its connectionId.- +41 more on the listing page
Ravendb MCP tools 21
list_compare_exchangeList cluster-wide compare-exchange (cmpxchg) key/value entries — used for atomic guards, unique constraints, and cluster-transaction state. Optional startsWith key prefix; paged (1-1024, default 100). Not document-scoped.get_database_statsPer-database statistics and state. Sections: Summary, Detailed, Collections, Indexing, IndexErrors, Storage, Tombstones, Metrics, Identities, Revisions, Sharding, HugeDocuments, Io. Choose with include; default is Summary + Collections + Indexing. For one index's deep performance/debug use get_index; for one storage environment's deep report use inspect_storage. Tombstones/Metrics/Sharding are availability-wrapped (e.g. Sharding only on sharded databases).collect_debug_packageDownload a RavenDB debug/support package (zip) to a local artifact file. scope=Server (this node), Cluster (cluster-wide), or Database (one database — needs databaseName). Returns the artifact path, content type, and byte size. Heavy; for deep support investigations.get_indexOne index, multiple views. Sections: Definition (maps/reduce, fields, config, lock/deployment mode), Staleness (is it stale and why), Debug (internal debug view + metadata + history), Terms (distinct indexed terms for fieldName, paged), Errors (this index's indexing errors), Performance (this index's performance stats). Choose with include; default is Definition + Staleness. Terms requires fieldName. For an all-indexes view use get_database_stats with the indexing section.run_queryRead-only RQL (no UPDATE/patch; page start/pageSize 1-128). Features | where: and or not in all-in between startsWith endsWith exists regex search boost fuzzy proximity exact lucene intersect cmpxchg id | order: cast score random custom | select: distinct projection js declare load | group: count sum key array | include: counters timeseries revisions highlight | spatial facet suggest morelikethis vector.search. Names only; for syntax read rql://index then the feature resource — don't write RQL from memory.sample_live_feedPull a live server feed for a few seconds and return what streamed in. feed selects the source: AdminLogs (operational logs), ClusterDashboard (throughput/requests/indexing/storage), TrafficWatch (HTTP/TCP requests as they happen — optional databaseName filter), GcEvents, Allocations, ThreadContention, or ThreadRunaway (a one-shot snapshot, ignores seconds). seconds is the capture window, 1-30. ThreadRunaway returns the top-5 threads by CPU in full plus a compact index of all threads; pass threadNamePrefix for full detail on specific threads. Returns the captured text with Truncated/Limit flags when capped.get_database_configConfiguration of one database. Sections: Settings (effective config keys), ClientConfig (client config pushed to clients), Studio, and feature toggles/policies Expiration, Refresh, DataArchival, Revisions, DocumentsCompression, TimeSeries, SchemaValidation. Choose with include; default is all. Feature sections are projected from the database record; null means not configured.get_network_detailsTCP/network details: Stats (server connection counts + bytes), Connections (active TCP connections — server-wide, or per database when databaseName is given), DatabaseInfo (endpoint a client/node uses to reach a database on a node — needs databaseName + nodeTag). Choose sections with include; default is Stats + Connections.get_database_recordDatabase record: topology, ongoing tasks (backup/replication/ETL/subscriptions), settings, and feature configuration. Index definitions and history are reduced to name+count (they dominate the raw record) — use get_index for one index's full definition/staleness/history.list_databasesList all database names in the cluster. Call first to discover targets.get_ai_agentsRavenDB AI Agents configured on a database — conversational agents (their model connection, system prompt, tools, parameters). This is the AI Agents feature, NOT GenAI embeddings/connection-string tasks. Omit name to list all agents; pass name for one agent's configuration. Availability-wrapped: reports unavailable when AI Agents isn't enabled or licensed.get_live_workloadLive runtime activity: running (and longest-running) operations, currently running queries plus the query cache, and transaction info. Choose sections with include (default: all). Pass operationId to fetch one operation's state instead of the overview. databaseName is required for Queries and (with operationId) Operations; for Operations/Transactions without it, results are server-wide.- +9 more on the listing page
Ready-to-paste client configs
Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
Mongodb MCP Server
{
"mcpServers": {
"mongodb-mcp-server": {
"command": "npx",
"args": [
"-y",
"mongodb-mcp-server@latest"
],
"env": {
"MDB_MCP_CONNECTION_STRING": "YOUR_MDB_MCP_CONNECTION_STRING_HERE"
}
}
}
}
Ravendb MCP
{
"mcpServers": {
"ravendb-mcp": {
"command": "npx",
"args": [
"-y",
"@ravendb/mcp"
],
"env": {
"RAVENDB_URLS": "YOUR_RAVENDB_URLS_HERE"
}
}
}
}
Frequently asked questions
Mongodb MCP Server is categorized under Databases and uses a local stdio subprocess. Ravendb MCP belongs to Databases and uses a local stdio subprocess. Pick Mongodb MCP Server for databases work and Ravendb MCP for databases.
Copy the mcpServers block from the config section on this page into your client's config (for example claude_desktop_config.json or ~/.cursor/mcp.json), then fully restart the client.
Mongodb MCP Server has no declared pricing — check the project's own site. Ravendb MCP has no declared pricing — check the project's own site. If a server needs environment variables such as API keys, define them under the env key in your client's MCP config.
Yes. MCP clients support multiple servers: add mongodb-mcp-server and ravendb-mcp as separate keys in the same mcpServers object, and your assistant will call tools from either one.
On AllMCPs, Mongodb MCP Server has a quality signal of 65/100 (Great) with 1 views, 0 installs and 1,146 GitHub stars. Ravendb MCP has 60/100 (Good) with 1 views, 0 installs and 2 GitHub stars.