Log inSubmit server
Google Knowledge Graph logo

Google Knowledge Graph

Searches Google’s Knowledge Graph for typed entities and returns structured metadata through two MCP tools.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Google Knowledge Graph

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "google-knowledge-graph": {
      "command": "npx",
      "args": [
        "-y",
        "@houtini/google-knowledge-graph-mcp"
      ],
      "env": {
        "GOOGLE_KNOWLEDGE_GRAPH_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

We couldn’t confirm it starts correctly

The install command started, but didn't respond the way we expected when we tried to talk to it.

npx -y @houtini/google-knowledge-graph-mcp

initialize succeeded but no response to tools/list.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 2d ago.

Overview

The google-knowledge-graph MCP server connects MCP clients to Google’s public Knowledge Graph Search API. It provides one tool for text-based entity searches and another for retrieving entities by Machine ID (MID), returning names, types, descriptions, links, images, and relevance scores as JSON. Use it for entity verification, resolution, or structured lookups involving people, places, organizations, concepts, and other supported entity types. It runs through NPX or a local Node.js build and requires a Google Cloud API key with the Knowledge Graph Search API enabled.

Use cases

  • Verify entity details during research
  • Resolve names to Google Machine IDs
  • Filter entities by schema.org type
  • Retrieve structured metadata for known entities

Key features

  • Searches Google Knowledge Graph entities
  • Looks up entities by Machine ID
  • Filters by language and entity type
  • Returns JSON entity metadata
  • Supports result limits up to 500

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 3d ago

Recent health check succeeded.

8/8 checks healthy over the last 45d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 3d ago via OSV.dev · @houtini/google-knowledge-graph-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars12Stargazers on the source repository.
npm downloads732Package downloads in the last 30 days.
Last commit17d agoMost recent push to the default branch.
Install checkInconclusiveStarted but did not complete the MCP handshake in our test.

How Google Knowledge Graph works

What google-knowledge-graph does

The google-knowledge-graph MCP server exposes Google’s Knowledge Graph Search API to MCP-compatible clients. It is intended for workflows that need structured information about real-world entities rather than a general web search. Typical targets include people, companies, locations, events, creative works, products, programming languages, sports teams, countries, and cities.

Search results can include an entity’s name, schema.org-style types, a short description, a more detailed description, an image URL, a related URL such as a Wikipedia page, a Machine ID, and a relevance score. Results are returned as JSON so the client can inspect or use the fields in a larger research or entity-resolution workflow.

How it works

The server communicates with the Knowledge Graph Search API endpoint at https://kgsearch.googleapis.com/v1/entities:search. Authentication uses an API key passed through the configured environment variable. The API response is JSON-LD, and the server presents the relevant entity data through MCP tools.

search_knowledge_graph accepts a required text query. Optional language codes and entity-type filters can narrow the search, while limit controls the number of results. The default limit is 20 and the documented maximum is 500. For example, a query for “Python” can be limited to the ComputerLanguage type.

lookup_knowledge_graph_entities accepts one or more Google Machine IDs, such as /m/0dl567, and can also receive language codes. This is useful when an earlier step has already identified the entity ID and a direct lookup is preferable to another name-based search.

Setup and configuration

The google-knowledge-graph MCP server can be launched with NPX using the @houtini/google-knowledge-graph-mcp package. Its required environment variable is GOOGLE_KNOWLEDGE_GRAPH_API_KEY.

To obtain credentials, create or select a project in Google Cloud, enable the Knowledge Graph Search API, and create an API key under Credentials. The documentation recommends restricting the key to that API. A Google Cloud key is required; a Google AI Studio key is not accepted. The API is described as free to use, subject to free-tier quotas.

Claude Desktop configuration uses an mcpServers entry with npx, the package name, and the API key in env. Claude Code uses claude mcp add instead and can be checked with claude mcp get. Local installations require a path to dist/index.js; Windows paths need escaped backslashes.

Tools and capabilities

  • Search for entities by name or descriptive query.
  • Filter searches by language and schema.org entity type.
  • Set a result limit from the default of 20 up to 500.
  • Retrieve entities directly by one or more Machine IDs.
  • Return names, types, descriptions, images, URLs, IDs, and relevance scores.

Limitations and notes

The google-knowledge-graph MCP server depends on Google Cloud API-key configuration and on the Knowledge Graph Search API being enabled. Invalid, expired, incorrectly formatted, or otherwise unsuitable keys produce authentication errors. Searches with overly narrow terms or type filters may return no entities; broadening the query or removing filters can help.

Free-tier quotas apply, with the README citing a general allowance of about 100,000 queries per day. The server uses CommonJS output for compatibility with Node.js MCP hosts. It does not provide general web search; its results come from Google’s Knowledge Graph entity dataset and the fields exposed by that API.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Google Knowledge Graph

google-knowledge-graph is an MCP server that connects MCP clients to Google’s public Knowledge Graph Search API. Its main tools search for entities by query and look up entities by Google Machine ID. Responses can include names, types, descriptions, images, URLs, relevance scores, and IDs for research or entity-resolution workflows.

Install it with NPX by configuring `npx -y @houtini/google-knowledge-graph-mcp` in an MCP client and setting `GOOGLE_KNOWLEDGE_GRAPH_API_KEY`. The README provides configuration examples for Claude Desktop and a separate `claude mcp add` command for Claude Code. A local source build is also supported.

Create a Google Cloud project, enable the Knowledge Graph Search API, and create an API key under Credentials. Set it as `GOOGLE_KNOWLEDGE_GRAPH_API_KEY`. The key must be a Google Cloud key; Google AI Studio keys are not supported. Restricting the key to the Knowledge Graph Search API is recommended.

It documents setup examples for Claude Desktop and Claude Code rather than Cursor specifically, so you must apply Cursor’s MCP configuration method and provide the required API-key environment variable.

Related MCP servers

All in Search & Data Extraction All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewGoogle Knowledge Graph AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/google-knowledge-graph?style=directory)](https://allmcps.com/mcp/google-knowledge-graph)
HTML Embed
<a href="https://allmcps.com/mcp/google-knowledge-graph"><img src="https://allmcps.com/api/badge/google-knowledge-graph?style=directory" alt="Google Knowledge Graph on AllMCPs" /></a>