Skip to the page
Get started free

SpecStream for GPT Researcher MCP retriever

SpecStream is the live web data engine for teams and agents. Add one entry to mcp_configs and set RETRIEVER=mcp, and GPT Researcher searches the live web and reads whole pages through SpecStream's MCP server.

Read the guide

What you get

One entry in mcp_configs
name, command, args and env, the four fields GPT Researcher reads for a local server. The key is SPECSTREAM_API_KEY.
Search and whole pages
The retriever calls search_web for results with their URLs and read_pages for the full text of pages, PDFs and Office files.
Alone or beside your retriever
RETRIEVER=mcp runs on SpecStream alone. RETRIEVER=tavily,mcp or searx,mcp keeps a web retriever for the sub queries.

Set it up

  1. In SpecStream, open Agents & API, create an API key with a spending limit and export it as SPECSTREAM_API_KEY.
  2. Run pip install specstream-mcp-stdio.
  3. Set RETRIEVER=mcp, pass the specstream entry in mcp_configs, and run your research.

The full setup guide has every step, with what to do if something doesn’t arrive.

Also in Agent frameworks & builders

The same live web data, in the other tools of this kind.

Live web data, in GPT Researcher MCP retriever.

Ask in your own words for companies, places, people, filings or anything else the web lists.

Read the guideTalk to a human