query_graphLABSCO SUMMARY
Built for the fact that a FalkorDB graph has no declared schema: the agent samples the shape before it writes a query.
The orientation path is the point — list the graphs, read the labels and relationship types, sample the properties that actually appear on them, and only then write Cypher. Property keys ranked by frequency also expose naming drift, which is the usual reason a hand-written query silently returns nothing.
What it is
FalkorDB's own server for its graph database: list the graphs, learn a schemaless graph's shape by sampling it, then run OpenCypher against it.
What you get
- OpenCypher execution against any graph on the instance, with a read-only variant that always runs through GRAPH.RO_QUERY
- Parameterised queries: values travel in a params object and are referenced as $name, instead of being concatenated into the Cypher text
- Graph schema discovery — node labels, relationship types, and optionally the connection topology as source, relationship and target triples
- Property-level discovery: sample nodes of a label or relationships of a type and get their property keys ranked by frequency, with the sampled count reported alongside what you asked for
- Listing the graphs on the instance, creating nodes and relationships, and deleting a graph you no longer want
- Read-only enforcement where it matters: all three schema tools run read-only, so they are safe against a replica
Requirements
A reachable FalkorDB instance and Node.js 18 or newer. Runs from npm as @falkordb/mcpserver, with FALKORDB_HOST and FALKORDB_PORT pointing at the database and FALKORDB_USERNAME and FALKORDB_PASSWORD when it is protected. Defaults to stdio; set MCP_TRANSPORT=http and MCP_PORT for a networked client, where MCP_API_KEY turns on API-key authentication. FALKORDB_DEFAULT_READONLY=true makes the whole server read-only, which is the setting for a replica.
Setup effort
One command — FALKORDB_HOST=localhost FALKORDB_PORT=6379 npx -y @falkordb/mcpserver
USE CASES
01
Querying a graph you did not build
List the graphs, read the labels and relationship types, then write the Cypher knowing what is really there.
02
Finding out why a query returns nothing
Sample the nodes of that label and see which property keys actually occur, and how often.
03
Reading from production without risking a write
Run through the read-only tool, which goes to GRAPH.RO_QUERY every time — safe against a replica.
04
Passing user input into a query safely
Send the values in a params object and reference them as $name in the Cypher, rather than building the query as a string.
05
Building up a graph as the conversation goes
Create the nodes and the relationships between them, then read back the schema to confirm the shape.
6 tools
Listed in project docs. Your client may show a different list.
query_graph_readonlylist_graphsget_graph_schemaget_node_schemaget_relationship_schemaBEFORE YOU INSTALL
What to know first
Tools you get
6 tools named in the project's documentation.
Requires
- npx on PATH
- Node.js 18+
Where it runs
Both documented setups run the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
FALKORDB_HOST=localhost FALKORDB_PORT=6379 npx -y @falkordb/mcpserverMCP Clients
Claude Desktop — npx
CLAUDE DESKTOP — NPX · JSON
{
"mcpServers": {
"falkordb": {
"command": "npx",
"args": [
"-y",
"@falkordb/mcpserver@latest"
],
"env": {
"FALKORDB_HOST": "localhost",
"FALKORDB_PORT": "6379",
"FALKORDB_USERNAME": "",
"FALKORDB_PASSWORD": ""
}
}
}
}Claude Desktop — local build
CLAUDE DESKTOP — LOCAL BUILD · JSON
{
"mcpServers": {
"falkordb": {
"command": "node",
"args": [
"<YOUR_PATH_TO_FALKORDB_MCPSERVER>/dist/index.js"
]
}
}
}
