Analysis updated 2026-06-21
Build a Discord or Telegram bot that can carry on conversations and take autonomous actions.
Create an AI agent that can browse the web, search documents, and respond to user queries across multiple chat platforms.
Deploy multiple coordinated AI agents that work together on different tasks, managed from a single web dashboard.
Scaffold and ship a custom AI agent product with its own user interface using the built-in project system.
| elizaos/eliza | teambit/bit | vercel-labs/skills | |
|---|---|---|---|
| Stars | 18,364 | 18,387 | 18,341 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | moderate | moderate | easy |
| Complexity | 3/5 | 4/5 | 2/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires API keys for your chosen AI model provider and credentials for whichever messaging platforms you connect.
Eliza (elizaOS) is an open source TypeScript framework for building and deploying autonomous AI agents. An "agent" in this context is software that can carry on conversations, perform tasks, and make decisions on its own, connected to various external services. The framework is organized around a few key concepts. The core runtime handles the agent's decision-making loop and memory. Plugins extend the runtime with specific behaviors, for example, there are plugins that connect an agent to messaging platforms like Discord and Telegram, plugins that give agents access to web browsing or document search, and plugins for working with various AI model providers. On top of this plugin system, developers can build full deployable products (called projects) that include their own user interface. Eliza comes with a command-line tool for scaffolding new agents and projects quickly, a web dashboard for managing agents in real time, and support for running multiple coordinated agents together. It also includes a set of ready-made example projects and evaluation tools for testing how well agents perform on various tasks. The project is designed to be model-agnostic, meaning the same agent code can run with different AI models from different providers by swapping configurations. It is licensed under the MIT license.
An open source TypeScript framework for building autonomous AI agents that can chat, make decisions, and connect to platforms like Discord and Telegram.
Mainly TypeScript. The stack also includes TypeScript, Node.js.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.