RushDB
Provides tools for interacting with RushDB, a property-centric graph database built on Neo4j. Enables ingesting JSON/CSV data, performing graph traversals, aggregations, vector similarity searches, and schema discovery through a unified JSON search interface.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@RushDB find all users who signed up in the last week"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
RushDB MCP Server
Give your agent a memory layer. Via MCP.
Connect RushDB to any MCP-compatible AI client and your agent gains persistent, structured memory — semantically searchable and graph-traversable, without managing embeddings or schemas.
Works with ChatGPT, Claude.ai, Claude Desktop, Cursor, Windsurf, Gemini CLI, and any MCP-compatible tool.
Two ways to connect
Remote OAuth (no installation)
For ChatGPT, Claude.ai, and other web-based AI clients — connect directly without installing anything locally:
MCP endpoint: https://mcp.rushdb.com/mcp
Your client redirects to RushDB for OAuth sign-in and receives a scoped token. No API key setup required.
ChatGPT:
Go to Settings → Connectors → Add connector
Enter
https://mcp.rushdb.com/mcpSign in with your RushDB account
Claude.ai:
Go to Settings → Integrations → Add integration
Enter
https://mcp.rushdb.com/mcpAuthorize with your RushDB account
Local (RushDB Cloud API key)
For Claude Desktop, Cursor, Windsurf, VS Code, and other stdio MCP clients:
1. Get an API key at app.rushdb.com
2. Add to your MCP client config:
{
"mcpServers": {
"rushdb": {
"command": "npx",
"args": ["@rushdb/mcp-server"],
"env": {
"RUSHDB_API_KEY": "your-api-key-here"
}
}
}
}Place this in:
Claude Desktop —
~/Library/Application Support/Claude/claude_desktop_config.jsonCursor — MCP settings in the Cursor config panel
Windsurf —
~/.codeium/windsurf/mcp_config.json
3. That's it. The agent now has 43 tools for storing, querying, searching and traversing structured memory.
RUSHDB_API_URLis optional and defaults tohttps://api.rushdb.com/api/v1, the RushDB Cloud API.
Local standalone engine (no Cloud account)
Run the Rust engine, then point a local stdio MCP client at it. This mode calls the public engine protocol directly; it does not use a RushDB Cloud API key, dashboard, or OAuth.
RUSHDB_TOKEN=local-development-token rushdb up --data-dir ./rushdb-data{
"mcpServers": {
"rushdb": {
"command": "npx",
"args": ["-y", "@rushdb/mcp-server"],
"env": {
"RUSHDB_MODE": "direct",
"RUSHDB_ENGINE_URL": "http://127.0.0.1:3000",
"RUSHDB_ENGINE_TOKEN": "local-development-token"
}
}
}
}Direct mode offers every tool whose work the engine does: schema discovery, record reads and writes (with
ordering, projections and aggregations), relationships, labels and properties, bulk import and CSV export, text
indexes and keyword search, and external vector indexes with vector search by a queryVector you supply.
The RushDB Cloud tools are not listed and are refused if called: smartSearch (natural-language search),
the relationship-pattern tools, and anything that needs managed embeddings (a text query for vector search,
managed indexes). The MCP HTTP transport is unavailable in direct mode because its OAuth flow belongs to
RushDB Cloud; run the server through stdio on the same machine as the client.
Related MCP server: Personal Neo4j Memory Server
What an agent can do
After connecting, ask Claude or your agent naturally:
"Remember that we decided to use Clerk for auth, replacing Auth0, on 2026-04-10."
The agent calls createRecord to store a structured memory. Later:
"What auth decisions have we made?"
The agent calls getSchemaMarkdown to learn what is stored, then findRecords, textSearch or vectorSearch.
"Show me everything related to our auth system."
The agent calls findRelationships and findRecords with relationship filters to traverse the graph of connected records.
Available tools
43 tools, named exactly as below. In direct mode the six RushDB Cloud tools are not listed. The full reference, with inputs and safety levels, is at docs.rushdb.com/3.x/learn/reference/mcp/tools.
Discovery and properties
Tool | What it does |
| The whole graph schema as compact Markdown; call once at session start |
| Structured schema JSON, with property IDs and vector indexes |
| List or filter record labels and their counts |
| Property names, types, IDs and record counts |
| Metadata for one property |
| Numeric or date ranges, or distinct string and boolean values |
| Remove a property and its values from all records |
| The complete SearchQuery syntax reference |
| The built-in query builder prompt (for clients without MCP Prompts support) |
Records
Tool | What it does |
| Search records; ordering, projections, grouping and aggregations |
| The first matching record |
| Exactly one matching record; fails on zero or several |
| One record by ID |
| Several records by ID |
| Insert or upsert one record |
| Update some fields of a record |
| Replace all fields of a record |
| Delete one record by ID |
| Insert or upsert many records with one label |
| Delete the records matching a query |
| Export matching records as CSV |
Relationships
Tool | What it does |
| Find relationships by type, properties or endpoint records |
| Create relationships between existing records |
| Remove relationships: every type unless |
Vector, keyword and hybrid search
Tool | What it does |
| List embedding indexes and their status |
| Create an embedding index (on a self-hosted engine, an external one with |
| How many records an index covers and has vectors for |
| Write your own vectors to an external index |
| Delete an embedding index and its vectors |
| Rank records by vector similarity after exact filtering; add |
| Deprecated alias for |
| List text indexes and their status |
| Declare a BM25 text index over one property, or |
| Drop a text index |
| Keyword (BM25) search after exact filtering |
RushDB Cloud only
Tool | What it does |
| Turn a natural-language request into a SearchQuery and run it |
| List inferred relationship patterns and analysis status |
| Queue schema analysis for relationship suggestions |
| Approve and apply a suggested pattern |
| Ignore a suggested pattern |
| Delete a pattern and optionally its materialized relationships |
Utilities
Tool | What it does |
| Setup instructions for your MCP client |
Writes accept an optional transactionId.
Built-in query builder prompt
RushDB exposes a system prompt via the MCP Prompts API that teaches the agent to query safely — discover labels and properties first, then construct queries.
How it works:
Client calls
prompts/list→ findsrushdb.queryBuilderClient calls
prompts/get→ injects it as the session system messageAgent now does discovery-first queries automatically
Most MCP clients handle this at session start. If your client doesn't support MCP Prompts, call getQueryBuilderPrompt and inject the result as your system message.
Environment variables
These apply to local (API key) mode. Remote OAuth connections do not use environment variables.
Variable | Required | Description |
| yes | Your RushDB API key |
| no | API base URL (default: |
For standalone direct mode, use these instead:
Variable | Required | Description |
| yes | Set to |
| yes | Standalone engine base URL, for example |
| yes | Token configured for the standalone engine. |
| no | Tenant on an engine started with |
Registry & autodiscovery
mcp.yamlat the repo root enables autodiscovery by MCP registriesglama.jsonat the repo root contains Glama metadataPackage:
@rushdb/mcp-server, command:npx
Development
git clone https://github.com/rush-db/mcp-server
cd mcp-server
npm install
npm run buildTroubleshooting: rushdb-mcp: command not found / MCP error -32000
This only happens when the MCP client launches the server with a working directory inside this repo (e.g. an .mcp.json at the repo root used by Claude Code / Cursor). This repo is @rushdb/mcp-server, so an unversioned npx @rushdb/mcp-server (even with -p) can resolve to the local package, whose rushdb-mcp bin isn't linked in node_modules/.bin. The process exits before the MCP handshake, which the client reports as -32000.
The published package is unaffected — the config above works as-is outside the repo. To run from inside the repo, pin a version so npx fetches from the registry instead of resolving the local package:
{
"mcpServers": {
"rushdb": {
"command": "npx",
"args": ["-y", "-p", "@rushdb/mcp-server@latest", "rushdb-mcp"],
"env": { "RUSHDB_API_KEY": "your-api-key-here" }
}
}
}Alternatively, run the local build directly (uses your in-repo code, must be built first with npm run build):
{
"mcpServers": {
"rushdb": {
"command": "node",
"args": ["build/index.js"],
"env": { "RUSHDB_API_KEY": "your-api-key-here" }
}
}
}License
Apache 2.0
Available Tools
29 toolsAttachRelationC
Create a relationship between records (single or multiple targets)
Inputs: sourceId (required), targetId, targetIds, relationType, +2 more.
BulkCreateRecordsC
Create multiple records in a single operation
Inputs: label (required), data (required), transactionId.
BulkDeleteRecordsC
Delete multiple records matching a query
Inputs: where (required), labels, transactionId.
CreateRecordC
Create a new record in the database
Inputs: label (required), data (required), transactionId.
DeletePropertyC
Delete a property from the database
Inputs: propertyId (required).
DeleteRecordC
Delete a record from the database (alias of DeleteRecordById)
Inputs: recordId (required), transactionId.
DeleteRecordByIdC
Delete a record by its ID
Inputs: recordId (required), transactionId.
DetachRelationB
Remove a relationship between records (single or multiple targets)
Inputs: sourceId (required), targetId, targetIds, relationType, +2 more.
ExportRecordsC
Export records to CSV format
Inputs: labels, where, limit, orderBy.
FindLabelsB
Find / filter record labels (supports where, limit, skip, orderBy). Superset of GetLabels.
Inputs: where, limit, skip, orderBy.
FindOneRecordC
Find a single record that matches the given search criteria
Inputs: labels, where.
FindPropertiesC
Find properties in the database using a search query
Inputs: where, limit, skip, orderBy.
FindPropertyByIdC
Find a specific property by ID
Inputs: propertyId (required).
FindRecordsC
Find records in the database using a search query
Inputs: labels, where, limit, skip, +3 more.
FindRelationshipsC
Find relationships in the database
Inputs: where, limit, skip, orderBy.
FindUniqRecordC
Find a unique record that matches the given search criteria
Inputs: labels, where.
GetQueryBuilderPromptA
Return the RushDB Query Builder system prompt. Use this if your MCP client does not support Prompts API.
Inputs: no named parameters.
GetRecordB
Get a specific record by ID
Inputs: recordId (required).
GetRecordsByIdsC
Get multiple records by their IDs
Inputs: recordIds (required).
GetSettingsB
Get the current database settings and configuration
Inputs: no named parameters.
HelpAddToClientA
Help the user add the RushDB MCP server to their MCP client
Inputs: no named parameters.
OpenBrowserC
Open a web browser to a specific URL
Inputs: url (required).
PropertyValuesC
Get values for a specific property
Inputs: propertyId (required), query, orderBy, limit, +1 more.
SetRecordC
Replace all fields of a record with provided values
Inputs: recordId (required), label (required), data (required), transactionId.
TransactionBeginB
Begin a new database transaction
Inputs: ttl.
TransactionCommitC
Commit a database transaction
Inputs: transactionId (required).
TransactionGetC
Get information about a transaction
Inputs: transactionId (required).
TransactionRollbackC
Rollback a database transaction
Inputs: transactionId (required).
UpdateRecordC
Update an existing record (partial update)
Inputs: recordId (required), label (required), data (required), transactionId.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 29 tools
Most tools have distinct purposes with clear boundaries, such as CreateRecord vs. UpdateRecord vs. DeleteRecord. However, there is some overlap between DeleteRecord and DeleteRecordById (described as aliases), and FindOneRecord vs. FindUniqRecord could be confusing without careful reading of descriptions, though the latter emphasizes uniqueness.
Tool names follow a consistent verb_noun pattern (e.g., CreateRecord, DeleteRecord, FindRecords) with clear action-object pairs. Minor deviations include GetQueryBuilderPrompt and HelpAddToClient, which are more descriptive but still maintain readability without disrupting the overall pattern.
With 29 tools, the count feels heavy for a database server, potentially overwhelming for agents. While it covers many operations, it could be streamlined by merging redundant tools like DeleteRecord and DeleteRecordById, or consolidating some find operations to reduce complexity.
The tool set provides comprehensive coverage for database operations, including CRUD (CreateRecord, GetRecord, UpdateRecord, DeleteRecord), bulk operations, relationships (AttachRelation, DetachRelation), transactions, and utilities like export and settings. No obvious gaps exist for core database workflows.
Related MCP Connectors
A private graph database you query in plain English or Cypher. Hosted, nothing to install.
Shared memory for AI agents, as a graph in your own Postgres. Writes never call an LLM.
Persistent knowledge graph for AI-augmented teams. Store decisions, findings, and standing rules across agent sessions with semantic search and typed connections. Includes cross-session memory, audit trail, workspace isolation, and secret detection. Built for teams running agents that need to remember. Free until launch with team tier as default, anon trial available.
A collaborative substrate over your data: vector, knowledge graph, SQL, geospatial, streaming.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceBuild Real-Time Knowledge Graphs for AI Agents32,946Apache 2.0
- AlicenseNot gradedqualityDmaintenanceProvides AI assistants with persistent, graph-based memory using Neo4j, enabling semantic search and complex relationship tracking. It features specialized Cloudflare Access support for secure remote connections and isolated multi-database management for different projects.7 npmMIT
- AlicenseNot gradedqualityNot gradedmaintenanceA lightning-fast, self-hosted knowledge store and memory layer for AI agents-
- AlicenseNot gradedqualityDmaintenanceProvides persistent knowledge graph memory for AI agents, enabling them to store, recall, and query facts about people, projects, and relationships across sessions.MIT