[ 01 / Reading the ranking ]
These tools solve different problems
OpenClaw and Hermes are autonomous execution agents. Paperclip and Gas Town help manage work across agent teams. Ruflo, CrewAI and LangGraph coordinate multiple agents. ECC, Superpowers and Spec Kit provide structured development practices, while LiteLLM and Mem0 offer infrastructure components.
This is a repository popularity ranking, not a verdict on which technology is best. GitHub stars indicate interest, not security, production readiness, maintenance quality or a verified benchmark advantage.
Each of the 56 profiles covers the project's role, typical use cases, a practical first experiment and limitations to consider. GitHub-derived facts are distinct from editorial recommendations.
Search and filter
Search by name or use case, filter by category, or choose alphabetical sorting.
Primary source links
Each profile links to the official GitHub repository with license and archival status.
Know the limits
Know what to verify before allowing agents to access tools, data, or production.
[ 02 / Ranking GitHub ]
56 AI projects
Sorted by GitHub stars, with a detailed profile for each project.
Results: 56 / 56
- #01OpenClaw ↗Autonomous agents
openclaw/openclaw
An open-source personal AI agent that can use tools and connected services to execute tasks beyond a single chat session.
392,062 GitHub - #02Superpowers ↗Workflows and practices
obra/superpowers
A collection of reusable development skills and structured workflows for agent-assisted software delivery.
299,452 GitHub - #04Hermes Agent ↗Autonomous agents
NousResearch/hermes-agent
Nous Research's extensible autonomous agent combines tools with persistent context and multi-step execution.
253,944 GitHub - #05OpenCode ↗Coding agents
anomalyco/opencode
An open-source coding agent focused on repository-aware work through terminal and developer workflows.
213,713 GitHub - #08Anthropic Agent Skills ↗Workflows and practices
anthropics/skills
Anthropic's official repository of reusable agent skills and task-specific instructions.
181,273 GitHub - #10Claude Code ↗Coding agents
anthropics/claude-code
Anthropic's coding-agent environment for working with repositories, tools, and iterative development tasks.
150,982 GitHub - #11LangChain ↗Multi-agent orchestration
langchain-ai/langchain
A modular application framework for composing LLMs, retrieval components, and agent tools.
147,861 GitHub - #12CC Switch ↗Memory and infrastructure
farion1231/cc-switch
A configuration manager for switching between coding-agent setups and model providers.
144,605 GitHub - #13Spec Kit ↗Workflows and practices
github/spec-kit
GitHub's toolkit for specification-driven development with AI coding assistants.
141,361 GitHub - #14OpenAI Codex ↗Coding agents
openai/codex
OpenAI's open-source coding-agent tooling for implementing and testing changes in repositories.
129,340 GitHub - #15Browser Use ↗Agent integrations
browser-use/browser-use
Browser automation tools designed to let AI agents interact with websites and browser sessions.
118,044 GitHub - #16Gemini CLI ↗Coding agents
google-gemini/gemini-cli
Google's command-line agent for code, repository analysis, and tool-assisted development.
107,383 GitHub - #17Paperclip ↗Agent work management
paperclipai/paperclip
An application for organizing AI agents, assigned work, and operational responsibility.
102,426 GitHub - #18OpenHands ↗Coding agents
OpenHands/OpenHands
A platform for software-engineering agents that execute tasks in code and tool environments.
90,953 GitHub - #19DeerFlow ↗Multi-agent orchestration
bytedance/deer-flow
ByteDance's open-source workflows for research and multi-step agent-powered deliverables.
83,899 GitHub - #21OpenSpec ↗Workflows and practices
Fission-AI/OpenSpec
A specification-driven workflow for describing and implementing software changes.
72,011 GitHub - #24Oh My OpenAgent ↗Multi-agent orchestration
code-yeongyu/oh-my-openagent
A workflow-oriented enhancement to OpenCode focused on coordinating coding agents.
70,056 GitHub - #26Microsoft AutoGen ↗Multi-agent orchestration
microsoft/autogen
Microsoft's framework for experimenting with message-based agent collaboration.
61,419 GitHub - #29OpenManus ↗Autonomous agents
FoundationAgents/OpenManus
An open-source project demonstrating approaches to general-purpose autonomous agents.
58,662 GitHub - #32BMAD Method ↗Workflows and practices
bmad-code-org/BMAD-METHOD
A role- and artifact-driven methodology for AI-assisted product and software delivery.
54,166 GitHub - #34LangGraph ↗Multi-agent orchestration
langchain-ai/langgraph
A state-graph framework for coordinating long-running agent workflows and checkpoints.
43,306 GitHub - #36wshobson/agents ↗Workflows and practices
wshobson/agents
A community collection of specialist agent prompts, skills, and reusable plugin patterns.
40,829 GitHub - #37Claude Code Router ↗Memory and infrastructure
musistudio/claude-code-router
An integration layer that helps route Claude Code requests to configured model providers.
37,689 GitHub - #38
reworkd/AgentGPT
A browser-based project exploring autonomous AI agents working toward user-defined objectives.
36,279 GitHub - #39AgentScope ↗Multi-agent orchestration
agentscope-ai/agentscope
An agent framework for building and evaluating single- and multi-agent applications.
33,305 GitHub - #40Composio ↗Agent integrations
ComposioHQ/composio
A toolkit of integrations that connects AI agents to external applications and actions.
30,560 GitHub - #41Deep Agents ↗Multi-agent orchestration
langchain-ai/deepagents
A framework for complex agent tasks involving tools, subagents, and extended workflows.
30,293 GitHub - #42OpenAI Agents SDK ↗Multi-agent orchestration
openai/openai-agents-python
OpenAI's Python SDK for agents, tools, and handoffs between specialists.
30,054 GitHub - #44Vibe Kanban ↗Agent work management
BloopAI/vibe-kanban
A work-management interface for delegating and monitoring coding-agent tasks.
28,372 GitHub - #45OpenAI Skills ↗Workflows and practices
openai/skills
An OpenAI skills repository containing example instructions for reusable agent tasks.
28,085 GitHub - #47OpenAI Symphony ↗Agent work management
openai/symphony
OpenAI's software-task orchestration project for running coding agents against a work queue.
27,696 GitHub - #49
RooCodeInc/Roo-Code
An editor-based coding-agent extension supporting multiple work modes and tools.
24,276 GitHub - #50Google ADK ↗Multi-agent orchestration
google/adk-python
Google's agent development kit for composing tool-using agents in Python.
21,862 GitHub - #51SWE-agent ↗Coding agents
SWE-agent/SWE-agent
A software-engineering agent focused on identifying and fixing repository issues.
20,570 GitHub - #52Gas Town ↗Agent work management
gastownhall/gastown
A multi-agent workspace for coordinating task ownership and persistent coding work.
18,376 GitHub - #53SuperAGI ↗Autonomous agents
TransformerOptimus/SuperAGI
A platform for experimenting with autonomous agent workflows and supporting tooling.
17,703 GitHub - #54Superset ↗Agent work management
superset-sh/superset
A workspace designed to run several coding-agent sessions in parallel using isolated working trees.
15,212 GitHub - #55Microsoft Agent Framework ↗Multi-agent orchestration
microsoft/agent-framework
Microsoft's framework for building applications with agents and orchestration.
14,140 GitHub - #56VoltAgent ↗Multi-agent orchestration
VoltAgent/voltagent
A TypeScript agent framework with support for tools and execution observability.
10,799 GitHub
[ 03 / Methods and sources ]
Where does the data come from?
Verified on October 15, 2026 via the GitHub REST API endpointGET /repos/{owner}/{repo}. Star counts, canonical names, archival status, licenses and last-push dates are from a dated GitHub API snapshot. Projects are ranked strictly by star count. These are historical observations, not counters refreshed at every page view.
The selection covers different areas of agent technology and is editorial, not an exhaustive index of GitHub. Stars do not establish security, software quality, production maturity or suitability for a specific workload. Use cases and suggested tests are editorial guidance, not results from an independent benchmark conducted by DARC.DEV.