MCP server · Security

Agent Trust Stack MCP server

Let your AI keep a tamper-evident log of what it did and check other agents' reputation scores.

This is a small helper program that gives your AI two record-keeping skills: a diary that cannot be quietly edited, and a way to give and read trust scores between AI agents. It is handy if you work with several AI agents and want to know what each one actually did, or whether another agent has a good track record. You do not need to be technical to use it, but it is aimed at people who already run AI agents.

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 something outside the chat. This helper gives your AI two skills: writing a tamper-evident activity log, and handling reputation scores between agents. Once it is connected, you just ask in plain words and the AI uses this helper behind the scenes.

What this MCP server does

You ask your AI to start a log or record something it just did. The AI calls this helper, which writes an entry into a file on your computer. Each entry is linked to the one before it with a fingerprint, so if anyone edits an old entry, the chain breaks and the check fails. You can also ask the AI to rate another agent, or to look up an agent's average score. When you ask for a check, the helper reads the files back and tells you whether everything still lines up.

Flow diagram: you ask your AI “Record that my agent finished reading 500 documents”, on your own computer the Agent Trust Stack MCP server works with your own computer, and you get back A yes or no, plus a short fingerprint.

Click to zoom

What you can do with it

Start a fresh activity log for an agent
Record things like decisions, results, errors, notes and milestones
Check whether the log has been tampered with
See stats about a log, such as length, agents and time span
Read the last few entries of a log
Give another agent a private score from -1.0 to 1.0
Look up an agent's average reputation score
Ask for external timestamp proof of a log

Try asking your AI

“Start a new activity log for my research agent”
“Record that the agent finished reading 500 documents today”
“Check whether the log has been changed since yesterday”
“Rate agent-b 0.8 for delivering accurate research”
“What is agent-b's reputation score right now”

What it gives back to you

You get short answers in the chat, like a confirmation that an entry was added, along with its position number and a short fingerprint. When you check a log, you get a simple yes or no plus a count of entries and which agents wrote them. When you look up a reputation, you get the number of ratings, the average score and how many different agents rated it. Nothing is hidden: the helper reads and writes plain files on your machine.

Before you start

What you need

Python installed on your computer
The agent-trust-stack-mcp package installed with pip
An MCP-compatible AI app such as Claude Code or Cursor

Good to know

The log and rating files live on your computer, so treat that folder as private, and remember that the optional timestamp step is the only time this helper talks to the internet.

Install it with your AI

Add Agent Trust Stack 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 Agent Trust Stack 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.