Frameworks
A code library that supplies the agent loop itself: orchestration, state, memory, tool calling and multi-agent coordination. You write your application inside its structure, and it is usually model-agnostic. AgentSidekick indexes 273 of them (0.37% of 73,820 products) and 248 have full profilesDER; the most notable are LangChain, RAGFlow and AutoGen. As of 2026-11-21.
Top frameworks
All in the directoryOrdered by notability: log of GitHub stars (sub-folder products weighted down), Open VSX downloads where present, plus fixed weight for researcher-checked records and organisation publishers. Not a quality score. Showing 50 of 248 profiled.
| # | Product | Publisher | Capabilities | GitHub stars | Last push | Licence | Record | Save and compare |
|---|---|---|---|---|---|---|---|---|
| 1 | LangChainFramework of model, tool, retriever and agent abstractions with hundreds of provider integrations; its agent API now runs on LangGraph. | LangChain | 150k | today | MIT | CUR | ||
| 2 | RAGFlowSelf-hostable RAG engine with deep document parsing, template-based chunking, citations and an agent builder. | infiniflow | 94k | 1d ago | Apache-2.0 | CUR | ||
| 3 | AutoGenFramework for multi-agent conversations and event-driven agent systems that popularised conversable-agent patterns. | microsoft | 62k | 9d ago | CC-BY-4.0 | CUR | ||
| 4 | CrewAIPython framework for role-based agent teams (crews) and event-driven Flows, independent of LangChain; paired with a commercial platform for deployment. | CrewAI | 61k | today | MIT | CUR | ||
| 5 | LlamaIndexFramework for building agents and workflows over your data, with connectors, indexing, retrieval and an event-driven Workflows engine. | LlamaIndex | 53k | 1d ago | MIT | CUR | ||
| 6 | LangGraphLow-level orchestration framework that models agents as stateful graphs with checkpointing, human-in-the-loop interrupts and durable execution. | LangChain | 45k | today | MIT | CUR | ||
| 7 | AgnoPython framework and runtime for agents and multi-agent teams with memory, knowledge and a production API server (AgentOS). | Agno | 43k | today | Apache-2.0 | CUR | ||
| 8 | DSPyFramework for programming LM pipelines as declarative modules with signatures, then compiling them with optimisers that tune prompts and weights against metrics. | Stanford NLP | 39k | 2d ago | MIT | CUR | ||
| 9 | AgentScopePython multi-agent framework with message-based agent communication, tool use, memory and a visual studio for debugging. | Alibaba (AgentScope) | 36k | 1d ago | Apache-2.0 | CUR | ||
| 10 | Deep AgentsAgent harness on LangGraph that adds planning tools, a virtual filesystem and subagent spawning for long-running, multi-step tasks. | LangChain | 31k | today | MIT | CUR | ||
| 11 | MastraTypeScript framework for agents, workflows, RAG, memory and evals, with a local dev playground and deploy targets for common hosts. | Mastra | 30k | today | Apache-2.0 | CUR | ||
| 12 | Semantic KernelSDK for integrating LLMs into .NET, Python and Java apps through plugins, function calling and agent abstractions. | microsoft | 29k | today | MIT | CUR | ||
| 13 | FastMCPPythonic framework for building MCP servers and clients with decorators, plus auth, proxying and server composition. | Prefect | 29k | today | Apache-2.0 | CUR | ||
| 14 | HaystackPython framework for composing pipelines and agents from components for retrieval, generation and tool use, aimed at production RAG and search. | deepset | 27k | today | Apache-2.0 | CUR | ||
| 15 | LettaPlatform for stateful agents with self-editing memory blocks, archival memory and a server that persists agents; originated as MemGPT. | Letta | 26k | 1mo ago | Apache-2.0 | CUR | ||
| 16 | Pydantic AIType-safe Python agent framework built around Pydantic validation, dependency injection, structured outputs and graph-based control flow. | pydantic | 21k | today | MIT | CUR | ||
| 17 | OpenAI EvalsFramework and registry of benchmarks for evaluating LLMs; OpenAI also offers a hosted Evals API in its platform. | openai | 20k | 7mo ago | MIT | CUR | ||
| 18 | CAMELMulti-agent framework for role-playing agent societies, synthetic data generation and large-scale agent simulations. | camel-ai | 18k | 5d ago | Apache-2.0 | CUR | ||
| 19 | PipecatPython framework for voice and multimodal conversational agents built as pipelines of streaming speech, LLM and TTS services. | Daily | 17k | today | BSD-2-Clause | CUR | ||
| 20 | LiveKit AgentsFramework for realtime voice and video agents that join LiveKit rooms, with turn detection and pluggable speech, LLM and TTS providers. | LiveKit | 15k | today | Apache-2.0 | CUR | ||
| 21 | VoltAgentTypeScript agent framework with memory, workflows, supervisor/sub-agent patterns and a companion observability console. | VoltAgent | 11k | 15d ago | MIT | CUR | ||
| 22 | GenkitOpen-source framework for AI features and agents with flows, tool calling, RAG and a local developer UI; works with Gemini and other providers. | 6.6k | today | Apache-2.0 | CUR | |||
| 23 | AG2Community fork of AutoGen 0.2 for multi-agent conversations, group chats and tool use, maintained independently of Microsoft. | ag2ai | 5.1k | today | Apache-2.0 | CUR | ||
| 24 | kagentKubernetes-native framework for defining, running and operating agents as custom resources, with built-in cloud-native tools. | CNCF (Solo.io) | 4.5k | today | Apache-2.0 | CUR | ||
| 25 | BeeAI FrameworkPython and TypeScript framework for production agents with requirement-based agents, workflows, memory and observability. | IBM (Linux Foundation) | 3.5k | 1d ago | Apache-2.0 | CUR | ||
| 26 | InspectPython framework for LLM and agent evaluations with datasets, solvers, scorers, sandboxed tool execution and a log viewer. | UK AI Security Institute | 3.3k | today | MIT | CUR | ||
| 27 | owl🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation | camel-ai | 20k | 5d ago | — | HEU | ||
| 28 | openfangOpen-source Agent Operating System | RightNow-AI | 18k | 4mo ago | Apache-2.0 | HEU | ||
| 29 | BettaFish微舆:人人可用的多Agent舆情分析助手,打破信息茧房,还原舆情原貌,预测未来走向,辅助决策!从0实现,不依赖任何框架。 | 666ghj | 43k | 1mo ago | GPL-2.0 | HEU | ||
| 30 | oh-my-claudecodeTeams-first Multi-agent orchestration for Claude Code | Yeachan-Heo | 41k | today | MIT | HEU | ||
| 31 | omnigentOmnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device. | omnigent-ai | 11k | today | Apache-2.0 | HEU | ||
| 32 | GenAI_Agents50+ tutorials and implementations for Generative AI Agent techniques, from basic conversational bots to complex multi-agent systems. | NirDiamant | 25k | 19d ago | — | HEU | ||
| 33 | adk-goAn open-source, code-first Go toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control. | 9.0k | 1d ago | Apache-2.0 | HEU | |||
| 34 | evolverThe GEP-powered self-evolving engine for AI agents. Auditable evolution with Genes, Capsules, and Events. | evomap.ai | EvoMap | 8.7k | 1d ago | GPL-3.0 | HEU | ||
| 35 | agents-towards-productionEnd-to-end, code-first tutorials for building production-grade GenAI agents. From prototype to enterprise deployment. | NirDiamant | 22k | 6d ago | — | HEU | ||
| 36 | MiroThinkerMiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively. | MiroMindAI | 8.5k | 4mo ago | Apache-2.0 | HEU | ||
| 37 | open-multi-agentSelf-hosted TypeScript agent runtime with durable approvals and verifiable run records. Own it, approve it, audit it. | open-multi-agent | 7.1k | 4d ago | MIT | HEU | ||
| 38 | MindSearch🔍 An LLM-based Multi-agent Framework of Web Search Engine (like Perplexity.ai Pro and SearchGPT) | InternLM | 6.9k | 1y ago | Apache-2.0 | HEU | ||
| 39 | agent-governance-toolkitAI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents. Covers 10/10 OWASP Agentic Top 10. | microsoft | 6.6k | today | MIT | HEU | ||
| 40 | intentkitIntentKit is an open-source, self-hosted cloud agent cluster that manages a collaborative team of AI agents for you. | crestalnetwork | 6.5k | 2mo ago | MIT | HEU | ||
| 41 | agentscope-javaBuild distributed, production-grade, long-running agents. | agentscope-ai | 6.4k | today | Apache-2.0 | HEU | ||
| 42 | RavenThe Harness of Harnesses • Super Agent: a trusted, persistent, self-evolving multi-agent ecosystem for all-domain collaboration. | EverMind-AI | 6.3k | today | Apache-2.0 | HEU | ||
| 43 | rllmDemocratizing Reinforcement Learning for LLMs | rllm-org | 5.9k | 1mo ago | Apache-2.0 | HEU | ||
| 44 | paper-orchestraPaperOrchestra is a multi-agent framework for automated AI research paper writing. | google-research | 5.6k | 6d ago | Apache-2.0 | HEU | ||
| 45 | skillhubSelf-hosted, open-source agent skill registry for enterprises. Publish & version skill packages, govern with RBAC and audit logs, deploy on-premise with Docker or Kubernetes. | iflytek | 5.4k | 1d ago | Apache-2.0 | HEU | ||
| 46 | solace-agent-meshPlatform for building and running AI agents across the full lifecycle — build, test, deploy, monitor, improve. Hosts your agents, runs their tools, and coordinates them over an event mesh, with sessions, artifacts, auth, evaluators, and observability built in. | SolaceProducts | 4.9k | 2d ago | — | HEU | ||
| 47 | koogKoog is a JVM (Java and Kotlin) framework for building predictable, fault-tolerant and enterprise-ready AI agents across all platforms – from backend services to Android and iOS, JVM, and even in-browser environments. Koog is based on our AI products expertise and provides proven solutions for complex LLM and AI problems | JetBrains | 4.7k | 11d ago | Apache-2.0 | HEU | ||
| 48 | youtu-agentA simple yet powerful agent framework that delivers with open-source models | TencentCloudADP | 4.6k | 12d ago | — | HEU | ||
| 49 | kungfuYour agents don’t hand off the work. Kungfu keeps the same Work moving across Codex, Claude, OpenCode, and your own execution surface. | kungfu-systems | 4.5k | today | Apache-2.0 | HEU | ||
| 50 | harnessrouterHarnessRouter Community Edition: the self-hosted, Apache-2.0 edition of the unified interface for agent harnesses. Run Codex, Claude Code, Hermes, PI, DSH, and more through one API, with sessions, streaming, files, cancellation, and failure handling. Implements the Unified Harness Protocol (UHP), an open standard. Your keys, your infrastructure. | HarnessRouter | 4.4k | today | Apache-2.0 | HEU |
Category mix
Top-level categories among frameworksDER
Whole catalogue: 256 classified of 273 indexed. A product can sit in several categories. As of 2026-11-21.
Show the data as a table
| products | |
|---|---|
| Agent Infrastructure | 242 |
| Developer Tools | 65 |
| Data & Knowledge | 63 |
| Cloud & Platform Integrations | 30 |
| Content & Media | 19 |
| Web & Browser | 13 |
| Productivity & Collaboration | 9 |
| Commerce & Finance | 6 |
| Business Automation | 6 |
Most common capabilities
Capabilities among frameworksDER
Whole catalogue, top 12 capabilities by product count. As of 2026-11-21.
Show the data as a table
| products | |
|---|---|
| Multi-agent Coordination | 66 |
| Prompt Optimization | 17 |
| Agent Memory | 15 |
| Python Agent Frameworks | 13 |
| Local Inference | 11 |
| Automated Research Reports | 10 |
| Voice Agents | 9 |
| Text to Speech | 8 |
| Multi-provider Access | 8 |
| Stateful Workflow Graphs | 5 |
| Policy Enforcement | 5 |
| Webhooks & Triggers | 4 |
Licence mix
Licences of frameworksSRC
Whole catalogue, 273 records. "None recorded" means no source states a licence, which is not the same as unlicensed.
Show the data as a table
| products | |
|---|---|
| MIT | 119 |
| Apache-2.0 | 90 |
| AGPL-3.0 | 6 |
| GPL-3.0 | 5 |
| CC-BY-4.0 | 2 |
| Unlicense | 1 |
| MPL-2.0 | 1 |
| GPL-2.0 | 1 |
| CC0-1.0 | 1 |
| Other licences | 1 |
| None recorded | 46 |
Launches per month
New frameworks per monthSRC
Whole catalogue, 207 records published 2024-11 to 2026-10. 12 records have no date and are not counted. The last month is partial.
Show the data as a table
| Frameworks (products) | |
|---|---|
| 2024-11 | 3 |
| 2024-12 | 4 |
| 2025-01 | 9 |
| 2025-02 | 3 |
| 2025-03 | 6 |
| 2025-04 | 5 |
| 2025-05 | 11 |
| 2025-06 | 5 |
| 2025-07 | 4 |
| 2025-08 | 8 |
| 2025-09 | 9 |
| 2025-10 | 4 |
| 2025-11 | 3 |
| 2025-12 | 4 |
| 2026-01 | 5 |
| 2026-02 | 21 |
| 2026-03 | 30 |
| 2026-04 | 18 |
| 2026-05 | 8 |
| 2026-06 | 11 |
| 2026-07 | 7 |
| 2026-08 | 17 |
| 2026-09 | 6 |
| 2026-10 | 6 |
Page data last updated .