ИИ-агенты
arcade-mcp
A Python framework for building MCP servers and agent tools.
★ 1,0 тыс. звёздMIT
Смотреть руководство →babyagi — проект с открытым исходным кодом из категории «ИИ-агенты» от yoheinakajima (yoheinakajima/babyagi на GitHub). У него 22 361 звезда на GitHub и 2 846 форков, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 31 января 2026.
Yohei Nakajima's autonomous task agent - now a self-building agent framework in Python
Decide what you want the agent to accomplish and which functions it will need.
Install the babyagi package from PyPI, or clone the repository to explore the source.
pip install babyagi, then register functions and store any key dependencies (such as API keys) the functions need.
Execute a registered function, watch the execution log, and try the self-building agent flow.
Build agents by registering Python functions that the agent can call to complete objectives.
Organize related functions into packs - including code-writing and self-build packs - and load them into an agent.
Store secrets and dependencies from code or through the dashboard so functions can use them at runtime.
Ships a self_build flow where the agent processes user input and writes new functions for itself.
Runs functions in an environment with a log you can inspect from code or the dashboard.
Includes a dashboard for managing functions and key dependencies.
Get the source code from GitHub and enter the project directory.
git clone https://github.com/yoheinakajima/babyagi.git
cd babyagiBabyAGI is published on PyPI, so a virtual environment plus pip is all you need.
pip install babyagiThe README's Basic Usage shows registering functions - including ones with dependencies - and executing them.
import babyagi
# register your functions, then execute one - see README Basic UsageIt was created by Yohei Nakajima and is maintained as an open-source project on GitHub.
Yes - BabyAGI is released under the MIT License, so it is free to use and modify.
The framework itself is free, but functions that call an LLM need your own provider API key (for example OpenAI), and usage is billed by that provider.
Compared with larger frameworks like LangChain or AutoGPT-style platforms, BabyAGI is deliberately small and function-centric - easier to read end to end, with a self-building twist, but fewer built-in integrations.
Run pip install babyagi, follow the README's Quick Start to register a function, and execute it - then explore the dashboard and the self-building agent example.
A Python framework for building MCP servers and agent tools.
Langroid: a principled Python framework for building LLM multi-agent applications (from CMU/UW researchers)
A framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.
Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat…
A frozen snapshot of the original BabyAGI autonomous-agent experiment.
Autonomous agent framework for Elixir: distributed, event-driven, dynamic.
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если babyagi вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к babyagi — продолжайте изучать полку по обе стороны от него.
Open-source LLM observability: trace, evaluate and monitor apps and agents
Stanford's Generative Agents simulation - 25 AI characters living in a virtual town
NVIDIA's reference stack for running AI agents more securely inside OpenShell sandboxes
Swarm — OpenAI's experimental framework for lightweight multi-agent orchestration with agents and handoffs, MIT
An MCP server that lets Claude, Cursor and Windsurf build n8n workflows for you
Google's code-first toolkit for building real AI agents.
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.