Agent frameworks & infrastructure

Agent frameworks & SDKs

12,596 agents · most starred first

Agent frameworks & SDKs, page 53 of 126

aiexe

A cutting-edge powerful AI agent

@damn-dev/cli

damn.dev — self-hosted workspace OS for human + AI agent collaboration.

mnemonica

AI-agent ready instance inheritance system for JavaScript/TypeScript via prototype chains

@open-gitagent/gitagent

A universal git-native multimodal always learning AI Agent (TinyHuman)

@stevederico/dotbot

AI agent CLI and library for Node.js — streaming, multi-provider, tool execution, autonomous tasks

@a5c-ai/agent-mux-core

Core types, client, and stream engine for agent-mux

marco-harness

Model-Agnostic Runtime for Controlled Orchestration — a small, readable AI agent harness.

@a5c-ai/agent-mux-adapters

Built-in agent adapters for agent-mux

@a5c-ai/agent-catalog

Public package for the shared agent ontology, discovery catalog, and packaged evidence graph consumed across the Babysitter toolchain.

@vibecontrols/vibe-plugin-ai-pi

Pi AI agent provider for VibeControls

ai-agent-flow

A Node.js framework for AI agent workflows

@ethosagent/cli

Ethos — TypeScript AI agent framework where personality is architecture. Curated toolsets, first-person identities, scoped memory.

create-scaffo

Create new scaffo AI agent harness SaaS projects

@vibecontrols/vibe-plugin-ai-openrouter

OpenRouter AI agent provider for VibeControls Agent (SDK)

aiden-runtime

Local-first autonomous work engine for Windows, WSL, and Linux with multiple providers and offline Ollama support.

@haklex/rich-agent-core

Headless AI agent protocol, diff engine, and store for Lexical editor

tavik

CLI for managing organization-owned AI agent guidance and reusable skills.

@tencent-ai/agent-sdk

CodeBuddy Code SDK for JavaScript/TypeScript

@mcp-abap-adt/llm-agent

Core interfaces, types, and lightweight default implementations for LLM agent orchestration.

@cleocode/lafs

LLM-Agent-First Specification schemas and conformance tooling

beeai-framework

BeeAI Framework - LLM Agent Framework

@jerryshell/pico-agent

一个轻量级 LLM Agent

@webmcp-auto-ui/agent

LLM agent loop + remote/WASM/local providers + MCP wrapper

@compilr-dev/agents

Lightweight multi-LLM agent library for building CLI AI assistants

@fluidframework/type-factory

Shared types for exposing schema methods and properties to an LLM agent

@mcp-abap-adt/sap-aicore-embedder

SAP AI Core embedding provider (IEmbedder) for @mcp-abap-adt/llm-agent.

@novu/agent-toolkit

Novu Agent Toolkit - expose Novu notification workflows as LLM agent tools.

@plurnk/plurnk-grammar

ANTLR4 grammar for the Plurnk LLM agent protocol

@mcp-abap-adt/ollama-embedder

Ollama embedding provider and OllamaRag convenience class for @mcp-abap-adt/llm-agent.

llmtxt

Primitives and SDK for LLM agent content workflows: compression, patching, progressive disclosure, signed URLs, collaborative document lifecycle, and retrieval planning

p-api-llm

一个轻量、高效、模型无关的 LLM Agent 框架。

ggai

OpenAI LLM Agent Interface

llm-agents

LLM Agent framework for Node.js

@studio-foundation/runner

Multi-provider LLM agent runner with tool plugin execution (Anthropic, OpenAI, mock). Used by Studio.

@snack-kit/porygon

Unified programmatic interface for LLM Agent CLI tools

@lewist9x/bee-agent-framework

Bee - LLM Agent Framework

@loopstack/agent

Generic LLM agent workflow for Loopstack — configurable agent loop with tool calling, error handling, and cancel support

agnostic-agents

A Node.js package for multi-LLM agent support

llm-agent

The llm-agent is a TypeScript project designed to facilitate interactions with language models from multiple providers. It supports various LLM API providers that are compatible with the OpenAI API structure, allowing users to manage chat interactions and

@tscg/tool-optimizer

High-level tool-schema optimizer for LLM agent frameworks. Integrates with LangChain, MCP, Vercel AI SDK.

@uncaged/workflow-agent-builtin

`uwf-builtin` agent — built-in LLM agent with file read/write and shell tools.

@mcp-abap-adt/llm-agent-libs

[![Stand With Ukraine](https://raw.githubusercontent.com/vshymanskyy/StandWithUkraine/main/badges/StandWithUkraine.svg)](https://stand-with-ukraine.pp.ua) [![License: LGPL v3](https://img.shields.io/badge/License-LGPL_v3-blue.svg)](https://www.gnu.org/lic

@mcp-abap-adt/llm-agent-server

Default SmartAgent implementation with LLM providers, MCP client, HTTP server, and CLI.

agentprofiles-cli

Manage configuration profiles for LLM agent tools

@mcp-abap-adt/llm-agent-mcp

[![Stand With Ukraine](https://raw.githubusercontent.com/vshymanskyy/StandWithUkraine/main/badges/StandWithUkraine.svg)](https://stand-with-ukraine.pp.ua) [![License: LGPL v3](https://img.shields.io/badge/License-LGPL_v3-blue.svg)](https://www.gnu.org/lic

@blindagency/browser

This package handles the browser side of the BlindAgency system: encrypting the user's API key before it leaves their device, managing the encrypted credential across requests, and running the LLM agent loop.

@earendil-works/pi-agent-core

General-purpose agent with transport abstraction, state management, and attachment support

hermes-handler

HermesHandler is a lightweight, framework-agnostic message router for browser extensions and event-driven systems, with strict {ok,result,error} envelopes, timeouts, cooperative cancellation, and safe normalization. Ideal for reliable LLM agent backbones.

create-llm-agent

Quickly scaffold a TypeScript-based LLM agent application with local or remote AI models

agent-base

Turn a function into an `http.Agent` instance

langchain

Typescript bindings for langchain

@hbbio/nanoagent

Lightweight, purely-functional LLM agent framework for the MCP era.

@mastra/core

Mastra is a framework for building AI-powered applications and agents with a modern TypeScript stack.

ottrix

TypeScript LLM agent framework: ReAct loop, tool calling, MCP, multi-agent workflows, memory, guardrails. Anthropic Claude, OpenAI, Ollama via fetch — vendor-neutral, zero SDK deps.

@convex-dev/agent

Convex component for AI agents with persistent chat threads, tool calling, streaming, and RAG.

p-api-agent

一个轻量、高效、模型无关的 LLM Agent 框架,采用 ReAct 单调用循环架构

@oh-my-pi/pi-agent-core

General-purpose agent with transport abstraction, state management, and attachment support

@fole/proteus

Provider-agnostic LLM agent framework. One LLMProvider interface, two adapters (Anthropic + OpenAI-compatible), tool loop, router/specialist orchestration, streaming, confirmation gate, typed errors, retry, channels. Zero runtime dependencies.

lll-web-agent

开箱即用的 LLM Agent SDK — 配个 API Key 就能跑

@f5xc-salesdemos/pi-agent-core

General-purpose agent with transport abstraction, state management, and attachment support

@fitalyagents/asynctools

Standalone async parallel tool execution for any LLM agent

@inngest/agent-kit

AgentKit is a framework for creating and orchestrating AI agents and AI workflows

@codluv/llmtxt

Core primitives for llmtxt: compression, validation, patching, progressive disclosure, signed URLs, and attachment helpers for LLM agent content workflows

@pmate/llm-agent

Standalone prompt agent powered by OpenAI SDK.

hermes-paperclip-adapter

Paperclip adapter for Hermes Agent — run Hermes as a managed employee in a Paperclip company

@elyracode/agent-core

General-purpose agent with transport abstraction, state management, and attachment support

@anthropic-ai/claude-agent-sdk

SDK for building AI agents with Claude Code's capabilities. Programmatically interact with Claude to build autonomous agents that can understand codebases, edit files, and execute workflows.

@eminent337/aery-core

General-purpose agent with transport abstraction, state management, and attachment support

@dyyz1993/pi-agent-core

General-purpose agent with transport abstraction, state management, and attachment support

@langchain/classic

Old abstractions from LangChain.js

@revealui/ai

AI runtime for agent-driven products — agents, memory, LLM providers (Inference Snaps, Ollama, OpenAI-compatible), tools, and orchestration. Anthropic-SDK-free.

esp-ai-plugin-llm-agent

ESP-AI plugin for LLM agent with streaming and non-streaming support

tunnel-agent

HTTP proxy tunneling agent. Formerly part of mikeal/request, now a standalone module.

beeai-framework-electron

BeeAI Framework - LLM Agent Framework

thepopebot

Create autonomous AI agents with a two-layer architecture: Next.js Event Handler + Docker Agent.

@amplitude/ua-parser-js

Detect Browser, Engine, OS, CPU, and Device type/model from User-Agent data. Supports browser & node.js environment

@codeparticle/libagent

Generic LLM agent framework with tool-based architecture

@llm-dev-ops/agentics-cli

Agentics CLI - Pure orchestration layer for agentics platform services

@azure/ai-agents

Azure AI Agents client library.

ts-llm-agent

Allows LLM to execute functions as part of the prompt tools

@flatfile/improv

A powerful TypeScript library for building AI agents with multi-threaded conversations, tool execution, and event handling capabilities

@idealeap/gwt

基于GWT理论构建的 LLM Agent 智能系统框架,不基于LangchainJs!

@npmcli/agent

the http/https agent used by the npm cli

http-cookie-agent

Allows cookies with every Node.js HTTP clients.

forever-agent

HTTP Agent that keeps socket connections alive between keep-alive requests. Formerly part of mikeal/request, now a standalone module.

@aws-sdk/util-user-agent-node

[![NPM version](https://img.shields.io/npm/v/@aws-sdk/util-user-agent-node/latest.svg)](https://www.npmjs.com/package/@aws-sdk/util-user-agent-node) [![NPM downloads](https://img.shields.io/npm/dm/@aws-sdk/util-user-agent-node.svg)](https://www.npmjs.com/

agentkeepalive

Missing keepalive http.Agent

@juliusbrussee/caveman-agent

General-purpose agent with transport abstraction, state management, and attachment support

@openeryc/pi-agent-core

General-purpose agent with transport abstraction, state management, and attachment support

default-user-agent

Default user agent string for nodejs http request

stratvibe

KISS substrate for LLM agent pipelines. One command. One job. No questions asked.

dotenv

Loads environment variables from .env file

@langchain/langgraph

LangGraph

@hoangvvo/llm-agent

The agent library to build LLM applications that work with any LLM providers.

toolpick

Dynamic tool selection for the Vercel AI SDK. Automatically picks the right tools per step.

@cleocode/core

CLEO core business logic kernel — tasks, sessions, memory, orchestration, lifecycle, with bundled SQLite store

llamaindex

<p align="center"> <img height="100" width="100" alt="LlamaIndex logo" src="https://ts.llamaindex.ai/square.svg" /> </p> <h1 align="center">LlamaIndex.TS</h1> <h3 align="center"> Data framework for your LLM application. </h3>

@framers/agentos

AgentOS: open-source TypeScript framework for autonomous AI agents. Unified graph orchestration, cognitive memory, runtime tool forging, multi-tier guardrails, voice pipeline, and 11 LLM providers.

@rozenite/agent-shared

Shared types and utilities for Rozenite agent packages

@cline/agents

Browser-safe agent runtime for the next-generation Cline SDK

More in Agent frameworks & infrastructure