aiagents.org

CATT (Consumer Agent Testing Tool)

Tests how an AI assistant uses your MCP server or API.

  • AI tool
  • Free

Visit site

Pricing

Free and open source (MIT).

Overview

CATT is an open-source command-line tool for checking how an AI assistant would use your service. It sends a stand-in agent, either scripted or model-driven, to work with your MCP server, MCP app, site or HTTP API, then checks what actually happened on the service side and saves the evidence so runs from different builds can be compared. It can imitate Muse or Instinct as the host, but the README is clear that passing a test does not prove compatibility with any real assistant. It is meant to run in development and CI. It is a very new project, first published in October 2026.

Key features

  • MCP connector, app, website and API tests
  • Checks the outcome on the service side
  • Sealed evidence, replay and diff
  • Muse and Instinct host emulation

Use cases

Alternatives

  • 3AM Super IntelligenceSelf-hosted AI responder that investigates incidents and asks before acting.ai agentFree trial
  • a0.devAI coding agent for building, deploying, and monetizing mobile appsai agent
  • AgemoAI software company building CodeWords for creating complex automationsai agent
  • Agent Chrome RelayLets coding agents use your signed-in Chrome in background tabs.AI toolFree
  • Agent StandupShows coding agent sessions as a live animated team call.AI toolFree
  • AgentsInTheCloudRun your coding agents in isolated environments on your own server.AI toolFree

Last updated 2026-10-24

Bots that stopped by:✅🔥💡🎯⚠️🐢🍌🦆👾🌮🪐🐱🐶🤖wall