Websets MCP connects AI assistants to Exa’s Websets API for building and enriching collections of web entities like companies, people, and research papers.

What you can do:

  • Find AI startups in San Francisco with funding over $10M
  • Build a database of companies and enrich with CEO names, revenue, employee counts
  • Create a list of research papers and extract key findings
  • Subscribe to webhooks to be notified when searches and enrichments complete
  • Bring your own CSV data into Websets via imports for scoping or exclusion

Installation

Get your Exa API key

Connect to Websets MCP:

https://websetsmcp.exa.ai/mcp?exaApiKey=YOUR_EXA_API_KEY
  • Cursor
  • VS Code
  • Claude Code
  • Claude Desktop
  • Windsurf
  • Other

Or add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "websets": {
      "url": "https://websetsmcp.exa.ai/mcp?exaApiKey=YOUR_EXA_API_KEY"
    }
  }
}

Run in terminal:

claude mcp add --transport http websets "https://websetsmcp.exa.ai/mcp?exaApiKey=YOUR_EXA_API_KEY"

Add to your Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "websets": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://websetsmcp.exa.ai/mcp?exaApiKey=YOUR_EXA_API_KEY"]
    }
  }
}

Available Tools

Webset management

Tool Description
create_webset Create a collection with search query, criteria, and enrichments
list_websets List all websets in your account
get_webset Get details about a webset
update_webset Update a webset’s title and/or metadata
delete_webset Delete a webset and all its items
preview_webset Preview how a query will be interpreted before creating a webset

Items

Tool Description
list_webset_items List items in a webset with their data
get_item Get details about a specific item

Searches

Tool Description
create_search Add more entities to a webset
get_search Check search status
cancel_search Cancel a running search

Enrichments

Tool Description
create_enrichment Extract data from items (text, number, date, email, phone, url, options)
get_enrichment Check enrichment status
cancel_enrichment Cancel a running enrichment

Webhooks

Tool Description
create_webhook Subscribe to real-time HTTP callbacks for webset events
get_webhook Get details about a webhook
update_webhook Update a webhook’s URL, events, or metadata
delete_webhook Delete a webhook
list_webhooks List all webhooks in your account

Imports

Tool Description
create_import Upload your own CSV data into Websets
get_import Get details about an import including its upload URL
list_imports List all imports in your account

Events

Tool Description
list_events List system events (search, enrichment, webset lifecycle, etc.)

Key Concepts

Entity Types: Search for different kinds of entities:

  • company - companies and startups
  • person - individuals (e.g., for recruiting)
  • research_paper - academic papers
  • article - blog posts and news articles
  • custom - define your own entity type

Criteria: Natural language filters that determine which results are included. Every result must satisfy all criteria to appear in your webset.

Enrichments: Data extractors that pull additional information from results that have already passed your criteria. Enrichments don’t affect filtering—they add columns of data to your results.

Enrichment Formats: The data type for an enrichment:

  • text - free-form text (CEO name, description)
  • number - numeric values (employee count, revenue)
  • date - dates (founding date, funding date)
  • email, phone, url - contact info
  • options - multiple choice (e.g., funding stage: Seed, Series A, Series B)

Example: To find AI startups and get their CEO names:

  • Criteria (filters): “AI company”, “Founded after 2020”, “Has raised funding”
  • Enrichments (data to extract): CEO name (text), funding amount (number), founding date (date)

Resources

Usage Examples

Create a Webset

Create a webset of AI startups in San Francisco founded after 2020. Find 10 companies and enrich with CEO name and funding amount.

Add More Entities

Search for 5 more AI companies that have raised Series A funding.

Extract Data

Add an enrichment to extract employee count for each company.

Subscribe to Events

Create a webhook to https://example.com/hook subscribed to webset.search.completed and webset.enrichment.completed events.