Using the Pursuit MCP to query PDF documents in signals
Last updated: June 22, 2026
The Pursuit MCP (Model Context Protocol) supports querying PDF documents that are embedded within opportunity signals.
See 📄 Using Pursuit with AI Assistants (MCP Integration) to get started.
Using the get-documents skill
To query PDF documents within signals, use the get-document tool in the Pursuit MCP. This tool allows you to:
Get insights into available documents
Query and search directly within documents
Retrieve the source URL for specific documents
Optimizing URL parsing for better performance
When providing signal URLs to the MCP agent, you can improve performance and reduce token usage by providing context about the URL structure. This helps the agent parse the correct inputs for get-document without needing to run search-documents first.
Use this URL structure format when working with the agent:
https://app.pursuit.us/account/<entityName>/<entityId>/doc/<urlId>https://app.pursuit.us/account/<entityName>/<entityId>/doc/<urlId>?page=<pageNumber>
Providing this context helps the agent understand the URL structure and directly use the appropriate parameters for the get-document tool.