Agent frameworks & infrastructure
Agent frameworks & SDKs
12,596 agents · most starred first
Agent frameworks & SDKs, page 112 of 126
Open-source Python SDK for embeddable, plugin-extensible agents
SDK for building BYO agents on decommons
Agent components for Dirac Compute.
A reusable Django package for building streaming AI agents with LangGraph and OpenRouter
Settings-agnostic Django agent-host substrate for Pydantic-AI: config in, built agent out.
Canonical location of the Dragonfly API definition
Expose Django REST Framework serializers and viewsets as OpenAI / MCP tools with automatic JSON Schema generation
Python base for DLLM-compatible conversational applications.
A plugin framework for agents, built on LangGraph deepagents + langgraph-cli.
A plugin framework for agents, built on LangGraph deepagents + langgraph-cli.
ECO AGENT - Five-layer loop driven autonomous AI agent system
An OpenAI-first agent harness core for controlled, durable, and observable agent runtimes
FableForge Agent — a fork of Hermes Agent (NousResearch). A self-improving AI agent that creates skills from experience, improves them during use, and runs anywhere
An opinionated FastAPI framework for AI/agent production applications
A typed FastAPI server compatible with LangGraph RemoteGraph clients
An OpenAPI-native API agent and embeddable FastAPI assistant.
A Python-first Agentic AI Framework — provider-agnostic, async-first, middleware-driven
Run Claude Agent SDK agents on Flyte.
Shared contract for Flyte agent-SDK adapters.
Run CrewAI agents on Flyte.
Run LangChain Deep Agents on Flyte.
Run Google ADK (Agent Development Kit) agents on Flyte.
Run Hermes agent on Flyte.
Run LangChain agents on Flyte.
Run LangGraph agents on Flyte.
Run Mistral Agents (Conversations API) on Flyte.
Run OpenAI Agents SDK agents on Flyte.
Run Pydantic AI agents on Flyte.
Async & Sync API client for FragAPI
A lightweight node-flow runtime for building tool-using agents. Import name: agent_core.
Decision-oriented agent runtime with explicit approvals, inspectable history, optional independent checking and a built-in web UI.
A placeholder package that reserves the name genericagentctrl.
Claude on Genkit
Ollama on Genkit
OpenAI-compatible models on Genkit
Declarative AI agent framework — MCP for tools, A2A for agents, AG-UI for humans
GEODE — 범용 자율 실행 에이전트
A glass-box, minimal library for building LLM agents — modern techniques as opt-in primitives, no hidden control flow.
Production runtime and network client SDK for Glyph
Google Cloud Agentidentity API client library
Google Cloud Agentregistry API client library
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Turn a prompt + a field schema into validated, typed JSON (Instructor over Gemini 2.5 Flash on Vertex AI).
Turn a prompt + a field schema into validated, typed JSON (Instructor over Gemini 2.5 Flash on Vertex AI).
Typed Python wrapper for deterministic gpt-workflow runs
Local agentic control plane for gpt2giga
Reserved alpha package for GraphBlocks
This name has been reserved using Reserver
Typed async client for the Hermes Agent API Server
HiveLLM binary RPC (Thunder) - wire v1 codec, family profiles, sync/async multiplexed clients
Hopeit integration for building agentic APIs
Shared runtime for the HPC MCP agent family: SSH middleware, PSI/J-style job models, scheduler backends, docs RAG, doctor, and serving, reused across per-machine plugins (Rikyu, HOKUSAI, Octopus, ...).
Portable agent runtime powered by the Hugpy self-hosted LLM fleet
An integration between iFlytek Spark and Haystack
Where fantasy meets functionality. The best of many worlds, refined into one intelligent agent.
inis.run CodeAct provider for Microsoft Agent Framework
A modular, pluggable agentive platform built on jvspatial
Deprecated — replaced by the 'kanari' package
Reusable runtime, config loader and bootstrap for KODA-ecosystem A2A agents on Amazon Bedrock AgentCore.
LangChain tools for dispatching reality-bound agent work to human operators.
A collection of tools for LangChain Mero.
OpenViking integrations for LangChain and LangGraph
RENAMED to 'langchain-tollwarden' (pip install langchain-tollwarden). Final compatibility release: re-exports the TollWarden LangChain integration under the old names and warns on import.
An integration package connecting Sarvam AI and LangChain
A provider-neutral Python client for LLM applications.
langgraph-agent-common
Declarative YAML workflow definitions for LangGraph — topology as config, logic in Python
Section-based flow control middleware for LangGraph React agents
Tiny, zero-dependency AI agent framework: tool-calling loop, function-to-schema tools, and memory over any OpenAI-compatible endpoint.
Anthropic Claude agent handler for LaunchDarkly AI SDK
Anthropic Claude messages handler for LaunchDarkly AI SDK
Alias stub for agents-live (take your agents live). Install agents-live instead.
Lightweight Python framework for building LLM agents with tool calling and RAG
LaC (LLM as Code) - reference engine. Declarative behavior for LLM apps: law in files, perimeter in code.
Run multi-stage LLM workflows: DAGs of prompts and scripts with automatic output chaining
Deterministic workflow topology enforcement for LLM-powered systems.
A minimal, ergonomic Python library for interacting with LLMs
Unified interface to call OpenAI and Anthropic-compatible LLM APIs from Python
A local-first AI orchestration framework
loopy-agent: 21 Essential AI Concepts in one toolkit — agentic loops, gateway, guardrails, evals, caching, observability, MCP, multi-agent, middleware, plugins, state, safety, cost, drift, skills, verification, audit, streaming, multi-modal, compliance, and explainability.
Pure Python ctypes wrapper for the Luxand LLM SDK
Maton Agent Toolkit
The self-improving AI agent — creates skills from experience, improves them during use, and runs anywhere
Entra ID Agent Container (sidecar) authentication library for Microsoft Agents
Integration library for Microsoft Agents with Teams
A dynamic, need-based multi-agent AI framework with session memory, action tracking, tool approval, and browser automation
Provider-neutral tool definitions and execution for Python agents
Server-side SDK for the MobileOPS Managed Agent API.
Composable Python runtime for building production AI agents
a kit for building LLM applications
A self-improving agent inspired by Hermes Agent.
Claude adapter for NeMo Fabric