TrikHub
TrikHub/Docs

TLDR;

For those in a hurry:

  • The entire goal of this project is to create a framework to allow anyone to create, publish and consume AI Agents (as we do with any other codebase, plugin, package, etc), safely;
  • There are concepts of Manifest and Passthrough to make sure developers audit and control the output of each Trik before using AND on runtime. Developers decide what they do with these but we make a clear separation of safe/risky;
  • Triks are “just code” with a auditable output. They can be as simple as a string return but meant to be full fledged AI Agents with their own dependencies, storage, env variables, flows etc;
  • Triks basically expose Actions which are automatically converted into Tools and exposed to your main Agent, ideally through LangGraph ;
  • Everything is kept under control through a Gateway;
  • Everything runs locally;
  • This project supports Python and Typescript/Node;

This project is in its infancy and many more updates, features, fixes and improvements are planned. Any help in the right direction is greatly appreciated. Check the Trikhub repository  for more info.

Core Concepts

Deep dive into TrikHub’s architecture and design principles. TrikHub is built on a simple principle: separate what the agent sees from what the user sees.

This section covers: