MCP server · Research
Agenda Intelligence MD
Your AI can check a research memo's evidence trail and flag broken references, quote mismatches, and gaps.
This is a small helper that checks the evidence behind an AI-written research memo. You give it the claims, the sources, the quotes, and the source text, and it tells you what does not line up. It is handy for analysts, researchers, and anyone who has to review a memo before it goes out.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill or a connection to an app or service. This one gives your AI a checking skill for evidence packets, so it can look at the claims and sources you hand it and report what is broken or missing. You still make the final call.
What this MCP server does
You give your AI a packet: a set of claims, the source IDs they point to, any quotes, and the source text itself. Your AI passes that packet to this helper. The helper runs a set of fixed checks: do the references resolve, do the quotes match the source text, do the numbers in the claims appear in the sources, and are there gaps in the evidence. It hands back a list of findings and a packet status, which your AI shows you in the chat. It does not go online to fetch missing sources, and it does not judge whether the sources are true.
Click to zoomWhat you can do with it
- Check whether every claim points to a source that is actually in the packet
- Compare declared quotes against the supplied source text
- Flag numbers in claims that do not appear in the sources
- List evidence gaps and reviewer actions to take next
- Run the same check from a command line or in a CI job
- Review a folder of documents and get an HTML report
Try asking your AI
- “Here is my evidence packet. Check it and tell me which references are broken.”
- “Do the quotes in this memo match the source text I pasted in?”
- “Which claims have numbers that are not in the sources?”
- “Give me the evidence gaps and what a reviewer should look at next.”
What it gives back to you
You get a packet status, a list of findings, and any evidence gaps or reviewer actions. Each finding points at the specific claim, reference, quote, or number that failed a check. In the chat this shows up as a short structured list or a JSON block you can read or pass on.
Before you start
What you need
- Python 3.9 or later
- The agenda-intelligence-md package installed
- An evidence packet file with claims, sources, quotes, and source text
Good to know
A packet can pass all checks and still be based on a stale or wrong source, so this is a preflight for human review, not proof that a claim is true.
Install it with your AI
Add Agenda Intelligence MD 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 Agenda Intelligence MD, 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
Analysts, researchers, editors, and compliance reviewers who need to inspect the evidence behind an AI-written memo before it is used.
