AI エージェント

オープンソースの AI エージェント

AI エージェントとは、言語モデルを使って次に何をするかを決め、ブラウザーやコードインタープリター、API などのツールを呼び出し、タスクが終わるまでそれを繰り返すプログラムです。ここには、独自のエージェントを作るためのフレームワークから、コードを書く・Web を閲覧する・業務フローを実行するといった完成済みのエージェントまでが揃っています。

選び方

比べるポイントは三つです。使えるモデル（特定の一社だけか、ローカルモデルを含む OpenAI 互換のエンドポイントなら何でも使えるか）、エージェントにできることをどう制限するか（サンドボックス、承認ステップ、権限リスト）、そしてステップ間やセッション間で記憶を保持できるか。コードやシェルコマンドを実行するエージェントはサンドボックスで動かすべきなので、自分のマシンへのアクセスを許可する前に必ず確認しましょう。

スターの多いプロジェクト

425 件中 409–425 件

polterguy

magic

★ 1,193

Instant SECURE Full Stack Apps and AI Agents

NVIDIA-NeMo

Gym

★ 1,189

Evaluate and improve models and agents using environments

Python

FunnyWolf

agentic-soc-platform

★ 1,181

Agentic SOC Platform: A powerful, flexible, open-source, and agent-centric automated security operations platform (AI SOC)

Python

rudrankriyam

Foundation-Models-Framework-Lab

★ 1,181

A practical lab for building, testing, and evaluating apps with Apple's Foundation Models framework.

Swift

AI-QL

tuui

★ 1,154

A desktop MCP client designed as a tool unitary utility integration, accelerating AI adoption through the Model Context Protocol (MCP) and enabling cross-vendor LLM API orchestration.

TypeScript

zjunlp

LightMem

★ 1,149

[ICLR 2026] LightMem: Lightweight and Efficient Memory-Augmented Generation

Python

Rath-Team

OpenRath

★ 1,136

An open-source, PyTorch-like runtime for dynamic multi-agent and multi-session workflows.

Python

massgen

MassGen

★ 1,130

🚀 MassGen is an open-source multi-agent scaling system that runs in your terminal, autonomously orchestrating frontier models and agents to collaborate, reason, and produce high-quality results. | Join us on Discord: discord.massgen.ai

Python

withkynam

vibecode-pro-max-kit

★ 1,130

Your AI forgets. This remembers. Spec-driven coding harness for vibecoders, product owners, CEOs and real builders — self-improving context memory, 15 agents, 33 skills working with /goal, agent-team, & workflow on autopilot loops with 0 need for human gate. Kills context rot, ships features, not spaghetti. Claude Code & Codex. Any stack

JavaScript

mrwadams

stride-gpt

★ 1,123

An AI-powered threat modeling tool that leverages OpenAI's GPT models to generate threat models for a given application based on the STRIDE methodology.

Python

vamplabAI

sgr-agent-core

★ 1,118

Schema-Guided Reasoning (SGR) has agentic system design created by neuraldeep community

Python

agentlas-ai

Agentlas-OS

★ 1,111

Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.

Python

Onelevenvy

flock

★ 1,109

A desktop multi-agent harness built with Rust, Tauri, and React, powered by langgraph-rust.

Rust

EmbeddedLLM

JamAIBase

★ 1,103

The collaborative spreadsheet for AI. Chain cells into powerful pipelines, experiment with prompts and models, and evaluate LLM responses in real-time. Work together seamlessly to build and iterate on AI applications.

Python

tigicion

dao-code

★ 1,084

Open-source TypeScript terminal coding agent for DeepSeek-V4 — builds on DeepSeek's strong price-performance and ultra-cheap cache pricing, engineering byte-stable prefixes and cache-reusing forks so cross-session memory and a continuous self-correction layer add almost no token cost; 1M context, Skills/MCP/Hooks, Claude Code config compatible.

TypeScript

raindrop-ai

workshop

★ 1,076

Give your coding agent the power to write and run agent evals.

TypeScript

Deuz-AI

Deuz-SDK

★ 988

Zero-dependency TypeScript framework for production AI agents: durable execution, long-term memory, hybrid RAG, MCP tool calling, human-in-the-loop approval, planning and CodeAct sandboxes. One streaming API for Claude, GPT, Gemini, Grok, Mistral and DeepSeek — Node, Bun, Deno, serverless and edge.

TypeScript