The Kanvo MCP Server: 61 Tools for Claude, Cursor and Windsurf

Connect your AI assistant or agent directly to Kanvo over the Model Context Protocol — read conversations, manage contacts, send attested messages, and inspect attestations.

What's in it

Conversations — list, search, read threads, mark read, archiveChat — Kanvo-native channels and direct messages: create, join, send, manage membersContacts — create, update, look up, organisationsReach — how a given person can be reached, and sending to them without naming a channelMessages — search across both stores,…

Authentication

Agents authenticate with a knv_ API key. Every key carries scopes — read-only by default — and the server only lists the tools the key's scopes reach, so a read-only key never even sees a send tool. Sending needs a key scoped for it and a mandate: an unexpired mandate covering that channel, or the call fails with a cl…

Connecting it

Add the server to your MCP client config:

Every agent send is attested

This is the part worth pausing on. When an agent calls the send tool, the same pipeline runs as everywhere else in Kanvo: resolve the actor, enforce the mandate, apply the disclosure notice, risk-score, deliver or divert to an approver, sign, append to the ledger.

Also available

If MCP is not your stack, the same surface is exposed as a REST API with inbound webhooks. Full reference on the API page.

Kanvo Team

Writes here about what Kanvo is building and why — the send path, the evidence it leaves, and the decisions behind both.

AI agents can now write, decide and act, but they still need somewhere to talk to people, with rules about what they may send. Here is what a messaging app built for agents needs, and how Kanvo gives your own agents, and other apps' agents, a place to work.

A Messaging App for AI Agents: Where Your Agents Talk to People

Enjoyed this article?

Subscribe to our newsletter for more insights and updates.