ИИ-агенты
voltagent
AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework
★ 10,7 тыс. звёздMIT
Смотреть руководство →langchainjs — проект с открытым исходным кодом из категории «ИИ-агенты» от langchain-ai (langchain-ai/langchainjs на GitHub). У него 18 248 звёзд на GitHub и 3 422 форка, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 8 октября 2026.
The JavaScript/TypeScript version of LangChain - the agent engineering platform
Decide the model, tools and data your agent or chain needs.
Install langchain from npm with npm, pnpm or yarn - no clone needed for normal use.
Install the integration package for your model provider and set its API key.
Compose a prompt, model and tool, run it, and iterate.
Call different LLM providers through one consistent interface and swap models with minimal code changes.
Tools, prompts, memory and retrievers composed into chains and agents.
First-party packages connect LangChain.js to major model providers, vector stores and tools.
Supported environments include Node.js and other JavaScript runtimes per the README.
Part of the LangChain ecosystem, so you can graduate to LangGraph agents and LangSmith observability.
Ships full TypeScript types for editor-friendly development.
Get the source code from GitHub and enter the project directory.
git clone https://github.com/langchain-ai/langchainjs.git
cd langchainjsInstall the core package with your package manager of choice.
npm install -S langchainImport from the package in your TypeScript app and follow the docs' first-agent examples.
import { ChatOpenAI } from "@langchain/openai"; // example provider package - see README resourcesIt is maintained by LangChain AI (langchain-ai) and contributors on GitHub.
Yes - LangChain.js is MIT licensed and free. Optional hosted services in the ecosystem (such as LangSmith) have their own pricing.
The framework is free, but the models you call usually need your own provider API keys; local model integrations are also available.
Compared with the Python LangChain, this is the same platform philosophy for JavaScript/TypeScript stacks; compared with writing raw provider SDK calls, it adds portable abstractions at the cost of another layer to learn.
Run npm install -S langchain, add your provider's integration package, and follow the documentation linked in the README.
AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework
The agent engineering platform
The batteries-included agent harness from the LangChain team.
YAO: rigorous engineering, evaluation, and governance for reusable agent skills.
JavaScript framework for multi-agent systems with a Kanban-inspired design.
iFLYTEK's enterprise agentic workflow platform — SuperAgents with workflows, RPA and one-click deploy.
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если langchainjs 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 langchainjs — продолжайте изучать полку по обе стороны от него.
Build customer-facing AI agents that actually follow your rules
An open-source agent operating system in Rust - autonomous 'Hands' that do real work
A self-healing harness that connects an LLM directly to your real browser
A production-grade platform for building agentic chat bots on every IM platform
Microsoft's trainer for AI agents - reinforcement learning for almost any agent framework
🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.