MCP server · Research
Corroborate MCP server
by chefcohen
Check how independently a news claim is being reported, with a confidence score and the sources behind it.

Corroborate is a small helper that checks how widely and independently a news claim is being reported. You give it a claim, and it tells you whether many separate outlets are covering it or whether it is really just one story echoed everywhere. It is handy for anyone who reads or shares news and wants a quick, honest read on the coverage before trusting it.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you using what it already knows. An MCP server is a small helper program that gives your AI a new skill or a connection to a service. This one connects your AI to public news sources like Google News, GDELT, and Hacker News. Once it is connected, your AI can look up how a claim is being reported and bring back a verdict with sources, without you doing the searching yourself.
What this MCP server does
You ask your AI something like whether a claim is being widely reported. Your AI passes the claim to this helper. The helper searches several public news engines at once, filters out articles that are not really about the claim, and groups near-identical headlines so that one wire story echoed by many outlets counts as one origin, not many. It then gives back a verdict, a confidence score from 0 to 1, and the list of sources it used. It also writes down its own caveats, like when a source engine was down or when it spotted heavy syndication.
Click to zoomWhat you can do with it
- Check whether a news claim is independently reported or just one story echoed around
- Get a confidence score from 0 to 1 for how well a claim is corroborated
- See the list of sources behind a verdict, with outlet, domain, and date
- Spot when a story comes from a wire service like AP or Reuters
- Ask for raw source lists without a verdict when you want to judge yourself
- See honest notes about syndication, breaking-news cascades, or engine outages
- Filter sources by how many days back to look, up to 90
Try asking your AI
- “Is it true that NASA delayed the Artemis III landing? How independently is that being reported?”
- “Check how corroborated this claim is: Acme Corp acquires Widget Industries for 2 billion dollars”
- “Find recent sources about the Federal Reserve holding interest rates steady”
- “Give me a corroboration verdict and confidence score for this headline I just saw”
What it gives back to you
You get back a verdict word like CONFIRMED, SINGLE_SOURCE, or UNCORROBORATED, plus a number of independent sources and a confidence score between 0 and 1. It also lists the sources it found, each with headline, outlet, web address, and date. There is a notes field with honest caveats, and a coverage field that says full or degraded if a search engine was down. If you use the find_sources tool instead, you just get the source list with no verdict.
Before you start
What you need
- Node.js 18 or newer on your computer
- An MCP client like Claude Desktop, Claude Code, or Cursor
- No accounts or API keys needed
Good to know
It measures how widely a claim is reported, not whether the claim is true, so a distorted version of a widely covered story can still come back CONFIRMED.
Install it with your AI
Add Corroborate MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check Corroborate MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.
Who it's for
Anyone who reads, writes, or shares news and wants a quick, honest read on how independently a claim is being reported, such as journalists, researchers, analysts, and curious office workers.





