Agent frameworks & infrastructure
Agent frameworks & SDKs
12,596 agents · most starred first
Agent frameworks & SDKs, page 109 of 126
Agentic Platform shared validation contract package
Typed, auditable action contracts for agentic software.
ACAIAgent — a model-agnostic layer for the Claude Agent SDK. Switch models mid-session without losing context. Bring your own models.
Agent402 for LangChain (the applied layer of Agentic Finance): turn Agent402's catalog of 500+ pay-per-call tools and skill packs (x402 or MPP) into LangChain tools for Python agents via four meta-tools. The pure-CPU tools are free via proof-of-work - no wallet, no API keys.
AI Agent 客制化配置中心 — 统一管理多 Agent 配置,一键启动
Program the Argus agent in your installed Cosmon app — drive SolidWorks, Abaqus, COMSOL, Fluent and more.
CrewAI integration for GaaS (Governance as a Service)
Pydantic AI integration for GaaS (Governance as a Service)
LangChain chat model provider for Apple's on-device Foundation Models framework
Typed Python client for integrating agents with meander.
Small uvx-based launcher for uv-agent plugin sets
Stateless, task-oriented AI agent core
AstriveAI - Universal AI Agent Framework
Sovereign AI Agent Desktop Daemon
Self-hosted, open-source AI agent platform — CLI
AI-Agent-One infrastructure agent CLI.
Engine 7 — Self-hosted AI agent engine
Scaffolds and orchestrates AI-agent development workspaces.
Production-ready TypeScript AI Agent Runtime
Kernova Core - The AI Agent Runtime Engine
Full-stack AI agent framework — frontend + agents in one project
Official @squasher-ai/agent package for Squasher.
Multi-frontend AI agent with full tool access, streaming, cron jobs, and plugin system
Multi-provider AI agent toolkit: generate language responses, search the web, extract content, and manage memory across 10+ LLM providers.
Runtime CLI for Lovrabet apps, datasets, SQL, Backend Function, skills, and AI agent workflows.
SUDO-AI - Autonomous AI Agent Platform
Composable AI agent workflow framework — execution, state, orchestration, runtimes, UI
Hyacinth 🪻 — multi-provider AI agent with tool system and context management
Multi-provider AI agent framework. Write agents once, deploy anywhere.
QuantaBit AI Agent SDK - Chat, tools, and session APIs for decentralized AI agents
AIGNE AI Agent exec provider — wraps runAgent/runCodeAgent as AFS exec actions at /dev/ai/agent
Self-hosted control plane for AI-agent workforces
Open Agent Pack — Declaratively manage AI agent infrastructure
Polpo SDK — typed HTTP client, SSE streaming, and reactive store for AI agent orchestration
AgentX — Decentralized AI Agent SDK with ReAct AgentLoop + Multi-Tenant LLM Providers
Drizzle ORM store implementations for the Polpo AI agent orchestration platform (PostgreSQL + SQLite)
Governed, durable-backed AI agent (chat + tool-calling + governance) for AdonisJS — part of the Agora ecosystem.
SuperSkill universal installer and compatibility CLI (hh) for reviewed AI-agent capabilities
Zero-Config AI Agent Portal Gateway & Multi-Stack Framework Bootstrapper
Self-hosted, self-evolving AI agent with its own dashboard.
Durable, provider-neutral AI agent harness for Cloudflare Workers
OpenAidy — self-hosted AI agent platform (prebuilt).
Governed, durable-backed AI agent for NestJS — @AiTool, discovery, /agent controllers
Provider-neutral AI agent runtime and MCP/Skill gateway for Marona Hub connections, managed agents, and bring-your-own-agent integrations.
Create a full-stack Kernova AI agent app
Astralform JS SDK - AI agent orchestration with WebMCP support
JavaScript/TypeScript SDK for AgentSkills Runtime - Install, manage, and execute AI agent skills with built-in runtime support
Micro Models Agent (MMA) — LLM agent harness for small models (Qwen3.5-9B, 32K-64K context)
Reusable LangGraph agent runtime: intent routing, sticky flows, Redis checkpointer, optional answer cache
Lightweight LLM agent framework built on node-llama-cpp — Agent, Memory, and Tool abstractions for local llama.cpp models
Neuralis — self-hosted, multi-LLM agent platform. Name reserved; `npm create neuralis` scaffolds the community host. See https://neuralisapp.com
Typed content blocks and message attribution for LLM agent output — one shape for chat transcripts and documents.
Gotong LLM agent base — provider interface and LlmAgent for building LLM-backed participants.
Тонкий шар роботи з LLM (локальні omlx + хмарні провайдери) поверх pi: model tiers, one-shot, agentic-раннери, write-guard, trace, telemetry, prompt-budget
Provider-neutral model adapter layer: canonical messages, capabilities, streaming contract
General-purpose agent with transport abstraction, state management, and attachment support
Rulvar core: L0 contracts, journal kernel, ctx primitives, agent runtime, model router, tool system, dynamic orchestrator, InMemory and JSONL stores, event stream.
General-purpose agent with transport abstraction, state management, and attachment support
Downcity Agent 运行时 — 单 Agent 执行壳与本机 RPC 能力
General-purpose agent with transport abstraction, state management, and attachment support
General-purpose agent with transport abstraction, state management, and attachment support
Rulvar first-class provider adapter for the OpenAI Responses API, plus the openaiCompatible factory.
Agent loop, tool runtime, event stream, context strategy, limits and cancellation
General-purpose agent with transport abstraction, state management, and attachment support
Mastra-backed AgentHarness implementation for @salesforce/sfdx-agent-sdk
General-purpose agent with transport abstraction, state management, and attachment support
Lightweight runtime library for brt-built botruntime agents
General-purpose agent with transport abstraction, state management, and attachment support
Probe any ACP-compatible agent for its capabilities (models, modes, configOptions, prompt capabilities, auth methods, MCP transports). No real prompt required; no token cost.
A TypeScript framework for building AI agent-powered applications. Integrates with multiple LLM providers (OpenAI, Anthropic, Ollama, Fireworks) and implements the AG-UI protocol for real-time frontend communication.
Zero Npm Dependencies Agent Framework
WAGENT Core - Gateway, Agent Engine, Memory
Rulvar flagship hybrid mode: plan agent, compileScript, WorkerSandboxRunner, self-repair loop. Plans before the run; not the PlanRunner orchestration extension, which is @rulvar/plan.
LangChain tool wrappers for the CRYPTYX x402 API. Drop-in DynamicStructuredTool instances that any LangChain agent can bind — signals, factor scores, institutional trigger evaluation, natural-language query. Payment plumbing handled via @cryptyx/x402-clie
Framework-agnostic Agent Protocol engine for LangGraph.js — run engine, handlers, and SSE, driven entirely by injected dependencies.
Framework-agnostic Agent Protocol engine for LangGraph.js — the heart of skein-js.
langgraph.json parser and graph loader (path:export) for skein-js.
Express adapter for skein-js — mount the Agent Protocol on an Express Router.
Next.js adapter for skein-js — serve the Agent Protocol from App Router or Pages Router API routes.
NestJS adapter for skein-js — serve the Agent Protocol from a Nest module.
Fastify adapter for skein-js — mount the Agent Protocol as a Fastify plugin.
Idiomatic NestJS integration for LangGraph: nodes as providers, typed edge lists, checkpointers, streaming, interrupts, and OpenTelemetry tracing
Framework-agnostic prebuilt LangChain tools for agents (works with any LangChain/LangGraph app)
Decorator-first, type-safe authoring layer for LangChain JS tools and agents.
Shared, framework-agnostic building blocks for skein-js HTTP adapters — in-memory dev runtime, LangGraph dev-state import, and langgraph.json CORS mapping.
Prebuilt mda binary for darwin/arm64.
Prebuilt mda binary for linux/x64.
LangChain integration for mekik — surface agent tool calls in the UI, gate them behind human approval, and make them exactly-once across an interrupt/resume.
Assembles skein-js ProtocolDeps from langgraph.json + selected drivers (memory/Postgres/Redis).
Prebuilt mda binary for win32/x64.
Prebuilt mda binary for darwin/x64.
Prebuilt mda binary for win32/arm64.
Prebuilt mda binary for linux/arm64.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Redis job queue and cross-instance pub/sub streaming for skein-js.
LangChain.js provider integration for SiliconFlow (chat + embeddings).
<div align="center">
AI agent packaging for HTTP QUERY — OpenAPI tool schemas, MCP, Vercel AI SDK, LangChain helpers.
A framework-agnostic TypeScript library for dynamic OpenAPI tool execution
Typescript bindings for langchain