MCP Server ​

▶ github.com/diffbot/diffbot-mcp

Plug Diffbot into any MCP client. The server exposes Diffbot's web fetching, web search, crawling, entity resolution, and Knowledge Graph search as tools your agent or LLM pipeline can call directly. It is the MCP counterpart to Agent Skills, for environments without bash access.

Compatible Clients

  • Claude Desktop and Claude Code
  • Visual Studio Code and GitHub Copilot
  • Cursor, Windsurf, and most IDEs that accept a standard mcpServers config
  • Any client that speaks MCP over stdio or HTTP

Installation ​

Quick Start ​

The easiest way to get started is to connect your MCP client to the remote server hosted by Diffbot. A free Diffbot token is required.

https://mcp.diffbot.com/mcp?token=<YOUR_DIFFBOT_TOKEN>

Authentication ​

The server looks for a token in this order:

  1. Authorization: Bearer <token> header or ?token=<token> query parameter (HTTP transport)
  2. DIFFBOT_TOKEN or DIFFBOT_API_TOKEN environment variable
  3. A DIFFBOT_API_TOKEN=<token> line in ~/.diffbot/credentials, the same file the db CLI and Agent Skills read, so one credential serves all three

Tools ​

Give your agent access to Diffbot's Knowledge Graph and web structuring APIs. Query over a trillion facts on organizations, people, and news; crawl and extract sites to build your own structured knowledge graphs.

Knowledge Graph ​

search_news ​

News and articles by mentioned entity, topic, publisher, language, date range, or sentiment. Newest first by default, relevance-ranked inside a date window.

search_organizations ​

Companies by industry, headquarters, headcount, revenue, funding, ownership, or leadership, plus similar_to lookalike lists for competitor mapping. Pass a name or a domain to look up a single company.

search_people ​

People by title, employer, employer industry, skills, education, location, or nationality. Covers people with a public online professional presence only — it is not a people-finder for private individuals.

search_places ​

Cities, counties, states, countries, and points of interest by name, population, prominence, containing place, or proximity.

search_deals ​

Funding rounds, investments, and acquisitions by date, size, series, investor, acquirer, or industry.

dql ​

The general-purpose layer under the five tools above, for any entity type or query shape they do not cover: products, patents, brands, job posts, cross-entity queries, and facet aggregations. Returns compact CSV rows with per-type default columns, or full JSON records on request. Every query can also be exported as a CSV download.

dql_probe ​

Parallel hit counts for several candidate queries. The way to check that a query is neither too broad nor too narrow before fetching results.

dql_ontology ​

Entity types, fields, taxonomies, and enums. A guessed field name is silently ignored by the graph, so this is how the agent confirms the exact spelling. Humans rarely need to touch it.

Web Structuring ​

search_web ​

Web search that ranks accuracy, not popularity, built on Diffbot's independent web index. Ranking heavily favors primary sources over secondary ones. Returns markdown content chunks for citation, with num_results and max_tokens controls.

Query url:<URL> to read Diffbot's already-parsed copy of a page from the index, in a fraction of the time of a live fetch.

extract ​

A web fetch alternative. Extract classifies the page it is fetching and returns content in a consistent JSON schema, optimized for minimal token usage while keeping data fidelity.

resolve_entities ​

Recognize organizations, people, places, and other proper nouns in text and resolve them to Knowledge Graph entity IDs. The response includes a ready-made dql_filter to fetch their full records with dql.

crawl ​

Find all links on a seed page and extract them all, with no scraping rules. Crawls run as background jobs and can be polled for status.

Example Prompts ​

# extract
→ Install Diffbot agent skills following these instructions https://www.diffbot.com/docs/interfaces/agent-skills

# crawl
→ Crawl Philips' websites for product data sheet PDFs

# search_organizations
→ AI companies based in San Francisco started by previous YC founders

# search_deals
→ recent M&As in pharma