# The Agentic Leaderboard

> The weekly-updated ranking of the top 100 open-source AI agents, scored on reliability, tool selection quality, autonomous iteration efficiency, and community mindshare. Last updated 2026-08-10T10:42:32Z. Total agents evaluated: 200.

## Methodology
- [Scoring formula and weights](https://www.theagenticleaderboard.com/methodology): Total = 0.35×Reliability + 0.20×Tool Selection Quality + 0.15×Autonomous Iteration + 0.10×Efficiency + 0.20×Community Mindshare
- [About + data sources](https://www.theagenticleaderboard.com/about)
- [Full leaderboard JSON](https://www.theagenticleaderboard.com/leaderboard.json)

## Top 10 Agents This Week
- [#1 cline](https://www.theagenticleaderboard.com/agent/cline): Score 87.7, category Coding & Software Engineering, by cline, repo https://github.com/cline/cline
- [#2 agentic-awesome-skills](https://www.theagenticleaderboard.com/agent/agentic-awesome-skills): Score 87.4, category Agent Framework, by sickn33, repo https://github.com/sickn33/agentic-awesome-skills
- [#3 ppt-master](https://www.theagenticleaderboard.com/agent/ppt-master): Score 87.3, category Autonomous Agent, by hugohe3, repo https://github.com/hugohe3/ppt-master
- [#4 ECC](https://www.theagenticleaderboard.com/agent/ecc): Score 87.0, category Agent Framework, by affaan-m, repo https://github.com/affaan-m/ECC
- [#5 AutoGPT](https://www.theagenticleaderboard.com/agent/autogpt): Score 87.0, category Autonomous Agent, by Significant-Gravitas, repo https://github.com/Significant-Gravitas/AutoGPT
- [#6 scientific-agent-skills](https://www.theagenticleaderboard.com/agent/scientific-agent-skills): Score 87.0, category Research & Reasoning, by K-Dense-AI, repo https://github.com/K-Dense-AI/scientific-agent-skills
- [#7 openclaw](https://www.theagenticleaderboard.com/agent/openclaw): Score 86.7, category General AI Agent, by openclaw, repo https://github.com/openclaw/openclaw
- [#8 n8n](https://www.theagenticleaderboard.com/agent/n8n): Score 86.6, category Workflow Orchestration, by n8n-io, repo https://github.com/n8n-io/n8n
- [#9 cherry-studio](https://www.theagenticleaderboard.com/agent/cherry-studio): Score 86.6, category Chatbot & Conversational, by CherryHQ, repo https://github.com/CherryHQ/cherry-studio
- [#10 dify](https://www.theagenticleaderboard.com/agent/dify): Score 86.5, category Workflow Orchestration, by langgenius, repo https://github.com/langgenius/dify

## All Top 100 Agents
- [#1 cline](https://www.theagenticleaderboard.com/agent/cline): Score 87.7
- [#2 agentic-awesome-skills](https://www.theagenticleaderboard.com/agent/agentic-awesome-skills): Score 87.4
- [#3 ppt-master](https://www.theagenticleaderboard.com/agent/ppt-master): Score 87.3
- [#4 ECC](https://www.theagenticleaderboard.com/agent/ecc): Score 87.0
- [#5 AutoGPT](https://www.theagenticleaderboard.com/agent/autogpt): Score 87.0
- [#6 scientific-agent-skills](https://www.theagenticleaderboard.com/agent/scientific-agent-skills): Score 87.0
- [#7 openclaw](https://www.theagenticleaderboard.com/agent/openclaw): Score 86.7
- [#8 n8n](https://www.theagenticleaderboard.com/agent/n8n): Score 86.6
- [#9 cherry-studio](https://www.theagenticleaderboard.com/agent/cherry-studio): Score 86.6
- [#10 dify](https://www.theagenticleaderboard.com/agent/dify): Score 86.5
- [#11 page-agent](https://www.theagenticleaderboard.com/agent/page-agent): Score 86.4
- [#12 langchain](https://www.theagenticleaderboard.com/agent/langchain): Score 86.3
- [#13 ai](https://www.theagenticleaderboard.com/agent/ai): Score 86.1
- [#14 DeepSeek-Reasonix](https://www.theagenticleaderboard.com/agent/deepseek-reasonix): Score 85.9
- [#15 jan](https://www.theagenticleaderboard.com/agent/jan): Score 85.8
- [#16 Agent-Reach](https://www.theagenticleaderboard.com/agent/agent-reach): Score 85.7
- [#17 career-ops](https://www.theagenticleaderboard.com/agent/career-ops): Score 85.6
- [#18 llama_index](https://www.theagenticleaderboard.com/agent/llama-index): Score 85.5
- [#19 last30days-skill](https://www.theagenticleaderboard.com/agent/last30days-skill): Score 85.5
- [#20 smolagents](https://www.theagenticleaderboard.com/agent/smolagents): Score 85.5
- [#21 cal.diy](https://www.theagenticleaderboard.com/agent/cal-diy): Score 85.4
- [#22 WrenAI](https://www.theagenticleaderboard.com/agent/wrenai): Score 85.4
- [#23 OpenMontage](https://www.theagenticleaderboard.com/agent/openmontage): Score 85.3
- [#24 agents](https://www.theagenticleaderboard.com/agent/agents): Score 85.3
- [#25 hermes-agent](https://www.theagenticleaderboard.com/agent/hermes-agent): Score 85.2
- [#26 learn-claude-code](https://www.theagenticleaderboard.com/agent/learn-claude-code): Score 85.2
- [#27 deer-flow](https://www.theagenticleaderboard.com/agent/deer-flow): Score 85.1
- [#28 crewAI](https://www.theagenticleaderboard.com/agent/crewai): Score 85.1
- [#29 OmniRoute](https://www.theagenticleaderboard.com/agent/omniroute): Score 85.1
- [#30 lobehub](https://www.theagenticleaderboard.com/agent/lobehub): Score 85.0
- [#31 openclaude](https://www.theagenticleaderboard.com/agent/openclaude): Score 85.0
- [#32 ToolJet](https://www.theagenticleaderboard.com/agent/tooljet): Score 84.9
- [#33 semantic-kernel](https://www.theagenticleaderboard.com/agent/semantic-kernel): Score 84.9
- [#34 sim](https://www.theagenticleaderboard.com/agent/sim): Score 84.9
- [#35 crush](https://www.theagenticleaderboard.com/agent/crush): Score 84.9
- [#36 langgraph](https://www.theagenticleaderboard.com/agent/langgraph): Score 84.8
- [#37 posthog](https://www.theagenticleaderboard.com/agent/posthog): Score 84.8
- [#38 dspy](https://www.theagenticleaderboard.com/agent/dspy): Score 84.8
- [#39 OpenCLI](https://www.theagenticleaderboard.com/agent/opencli): Score 84.8
- [#40 deepagents](https://www.theagenticleaderboard.com/agent/deepagents): Score 84.7
- [#41 kestra](https://www.theagenticleaderboard.com/agent/kestra): Score 84.7
- [#42 cc-haha](https://www.theagenticleaderboard.com/agent/cc-haha): Score 84.7
- [#43 pydantic-ai](https://www.theagenticleaderboard.com/agent/pydantic-ai): Score 84.7
- [#44 cognee](https://www.theagenticleaderboard.com/agent/cognee): Score 84.6
- [#45 budibase](https://www.theagenticleaderboard.com/agent/budibase): Score 84.6
- [#46 ruflo](https://www.theagenticleaderboard.com/agent/ruflo): Score 84.5
- [#47 anything-llm](https://www.theagenticleaderboard.com/agent/anything-llm): Score 84.5
- [#48 signoz](https://www.theagenticleaderboard.com/agent/signoz): Score 84.5
- [#49 headroom](https://www.theagenticleaderboard.com/agent/headroom): Score 84.4
- [#50 ai-engineering-from-scratch](https://www.theagenticleaderboard.com/agent/ai-engineering-from-scratch): Score 84.4
- [#51 gemini-cli](https://www.theagenticleaderboard.com/agent/gemini-cli): Score 84.3
- [#52 agent-skills](https://www.theagenticleaderboard.com/agent/agent-skills): Score 84.3
- [#53 nanoclaw](https://www.theagenticleaderboard.com/agent/nanoclaw): Score 84.2
- [#54 daily_stock_analysis](https://www.theagenticleaderboard.com/agent/daily-stock-analysis): Score 84.1
- [#55 goose](https://www.theagenticleaderboard.com/agent/goose): Score 84.1
- [#56 tidb](https://www.theagenticleaderboard.com/agent/tidb): Score 84.1
- [#57 Vibe-Trading](https://www.theagenticleaderboard.com/agent/vibe-trading): Score 84.1
- [#58 gpt-researcher](https://www.theagenticleaderboard.com/agent/gpt-researcher): Score 84.1
- [#59 ollama](https://www.theagenticleaderboard.com/agent/ollama): Score 84.0
- [#60 langflow](https://www.theagenticleaderboard.com/agent/langflow): Score 84.0
- [#61 awesome-copilot](https://www.theagenticleaderboard.com/agent/awesome-copilot): Score 84.0
- [#62 hyperframes](https://www.theagenticleaderboard.com/agent/hyperframes): Score 84.0
- [#63 composio](https://www.theagenticleaderboard.com/agent/composio): Score 84.0
- [#64 edict](https://www.theagenticleaderboard.com/agent/edict): Score 84.0
- [#65 ponytail](https://www.theagenticleaderboard.com/agent/ponytail): Score 83.9
- [#66 mem0](https://www.theagenticleaderboard.com/agent/mem0): Score 83.9
- [#67 openinterpreter](https://www.theagenticleaderboard.com/agent/openinterpreter): Score 83.9
- [#68 zeroclaw](https://www.theagenticleaderboard.com/agent/zeroclaw): Score 83.9
- [#69 openhuman](https://www.theagenticleaderboard.com/agent/openhuman): Score 83.9
- [#70 planning-with-files](https://www.theagenticleaderboard.com/agent/planning-with-files): Score 83.9
- [#71 UI-TARS-desktop](https://www.theagenticleaderboard.com/agent/ui-tars-desktop): Score 83.8
- [#72 firecrawl](https://www.theagenticleaderboard.com/agent/firecrawl): Score 83.7
- [#73 claude-mem](https://www.theagenticleaderboard.com/agent/claude-mem): Score 83.7
- [#74 activepieces](https://www.theagenticleaderboard.com/agent/activepieces): Score 83.6
- [#75 kilocode](https://www.theagenticleaderboard.com/agent/kilocode): Score 83.6
- [#76 FinceptTerminal](https://www.theagenticleaderboard.com/agent/finceptterminal): Score 83.6
- [#77 CowAgent](https://www.theagenticleaderboard.com/agent/cowagent): Score 83.5
- [#78 conductor](https://www.theagenticleaderboard.com/agent/conductor): Score 83.5
- [#79 agency-agents](https://www.theagenticleaderboard.com/agent/agency-agents): Score 83.4
- [#80 pi](https://www.theagenticleaderboard.com/agent/pi): Score 83.4
- [#81 LibreChat](https://www.theagenticleaderboard.com/agent/librechat): Score 83.4
- [#82 NemoClaw](https://www.theagenticleaderboard.com/agent/nemoclaw): Score 83.4
- [#83 ragflow](https://www.theagenticleaderboard.com/agent/ragflow): Score 83.3
- [#84 open-design](https://www.theagenticleaderboard.com/agent/open-design): Score 83.3
- [#85 oh-my-openagent](https://www.theagenticleaderboard.com/agent/oh-my-openagent): Score 83.3
- [#86 qlib](https://www.theagenticleaderboard.com/agent/qlib): Score 83.3
- [#87 adk-python](https://www.theagenticleaderboard.com/agent/adk-python): Score 83.3
- [#88 agentmemory](https://www.theagenticleaderboard.com/agent/agentmemory): Score 83.3
- [#89 jcode](https://www.theagenticleaderboard.com/agent/jcode): Score 83.3
- [#90 continue](https://www.theagenticleaderboard.com/agent/continue): Score 83.2
- [#91 mlflow](https://www.theagenticleaderboard.com/agent/mlflow): Score 83.2
- [#92 openai-agents-python](https://www.theagenticleaderboard.com/agent/openai-agents-python): Score 83.2
- [#93 oh-my-pi](https://www.theagenticleaderboard.com/agent/oh-my-pi): Score 83.2
- [#94 hive](https://www.theagenticleaderboard.com/agent/hive): Score 83.2
- [#95 OpenHands](https://www.theagenticleaderboard.com/agent/openhands): Score 83.1
- [#96 agno](https://www.theagenticleaderboard.com/agent/agno): Score 83.1
- [#97 agentscope](https://www.theagenticleaderboard.com/agent/agentscope): Score 83.1
- [#98 SWE-agent](https://www.theagenticleaderboard.com/agent/swe-agent): Score 83.1
- [#99 agent-framework](https://www.theagenticleaderboard.com/agent/agent-framework): Score 83.1
- [#100 OpenViking](https://www.theagenticleaderboard.com/agent/openviking): Score 82.9

## Categories
- [Coding & Software Engineering](https://www.theagenticleaderboard.com/category/coding-software-engineering)
- [Agent Framework](https://www.theagenticleaderboard.com/category/agent-framework)
- [Autonomous Agent](https://www.theagenticleaderboard.com/category/autonomous-agent)
- [Browser & Computer Use](https://www.theagenticleaderboard.com/category/browser-computer-use)
- [Research & Reasoning](https://www.theagenticleaderboard.com/category/research-reasoning)
- [Workflow Orchestration](https://www.theagenticleaderboard.com/category/workflow-orchestration)
- [Multi-Modal Agent](https://www.theagenticleaderboard.com/category/multi-modal-agent)

## Weekly Archive
- [Weekly digests](https://www.theagenticleaderboard.com/weekly)

## How to cite
The Agentic Leaderboard, weekly snapshot. theagenticleaderboard.com. Accessed [date].

---

## Full Agent Details

### #1 cline
- URL: https://www.theagenticleaderboard.com/agent/cline
- Developer: cline
- Category: Coding & Software Engineering
- Repository: https://github.com/cline/cline
- Overall score: 87.7 / 100
- Description: Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 65950
- Community mindshare score: 89.31

### #2 agentic-awesome-skills
- URL: https://www.theagenticleaderboard.com/agent/agentic-awesome-skills
- Developer: sickn33
- Category: Agent Framework
- Repository: https://github.com/sickn33/agentic-awesome-skills
- Overall score: 87.4 / 100
- Description: AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 2,005+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 44716
- Community mindshare score: 87.76

### #3 ppt-master
- URL: https://www.theagenticleaderboard.com/agent/ppt-master
- Developer: hugohe3
- Category: Autonomous Agent
- Repository: https://github.com/hugohe3/ppt-master
- Overall score: 87.3 / 100
- Description: AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and support for your own .pptx templates. · by Hugo He
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 44259
- Community mindshare score: 86.68

### #4 ECC
- URL: https://www.theagenticleaderboard.com/agent/ecc
- Developer: affaan-m
- Category: Agent Framework
- Repository: https://github.com/affaan-m/ECC
- Overall score: 87.0 / 100
- Description: The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 239111
- Community mindshare score: 96.85

### #5 AutoGPT
- URL: https://www.theagenticleaderboard.com/agent/autogpt
- Developer: Significant-Gravitas
- Category: Autonomous Agent
- Repository: https://github.com/Significant-Gravitas/AutoGPT
- Overall score: 87.0 / 100
- Description: AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
- Reliability success rate: 100.0%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 186483
- Community mindshare score: 96.37

### #6 scientific-agent-skills
- URL: https://www.theagenticleaderboard.com/agent/scientific-agent-skills
- Developer: K-Dense-AI
- Category: Research & Reasoning
- Repository: https://github.com/K-Dense-AI/scientific-agent-skills
- Overall score: 87.0 / 100
- Description: Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 158 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and t
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 33108
- Community mindshare score: 85.44

### #7 openclaw
- URL: https://www.theagenticleaderboard.com/agent/openclaw
- Developer: openclaw
- Category: General AI Agent
- Repository: https://github.com/openclaw/openclaw
- Overall score: 86.7 / 100
- Description: Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞 
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 385748
- Community mindshare score: 100.0

### #8 n8n
- URL: https://www.theagenticleaderboard.com/agent/n8n
- Developer: n8n-io
- Category: Workflow Orchestration
- Repository: https://github.com/n8n-io/n8n
- Overall score: 86.6 / 100
- Description: Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
- Reliability success rate: 100.0%
- Tool selection quality: 69.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 200056
- Community mindshare score: 97.1

### #9 cherry-studio
- URL: https://www.theagenticleaderboard.com/agent/cherry-studio
- Developer: CherryHQ
- Category: Chatbot & Conversational
- Repository: https://github.com/CherryHQ/cherry-studio
- Overall score: 86.6 / 100
- Description: AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 50216
- Community mindshare score: 87.61

### #10 dify
- URL: https://www.theagenticleaderboard.com/agent/dify
- Developer: langgenius
- Category: Workflow Orchestration
- Repository: https://github.com/langgenius/dify
- Overall score: 86.5 / 100
- Description: Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 151942
- Community mindshare score: 94.48

### #11 page-agent
- URL: https://www.theagenticleaderboard.com/agent/page-agent
- Developer: alibaba
- Category: Browser & Computer Use
- Repository: https://github.com/alibaba/page-agent
- Overall score: 86.4 / 100
- Description: JavaScript in-page GUI agent. Control web interfaces with natural language.
- Reliability success rate: 100.0%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 28563
- Community mindshare score: 84.46

### #12 langchain
- URL: https://www.theagenticleaderboard.com/agent/langchain
- Developer: langchain-ai
- Category: Agent Framework
- Repository: https://github.com/langchain-ai/langchain
- Overall score: 86.3 / 100
- Description: The agent engineering platform.
- Reliability success rate: 100.0%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 143868
- Community mindshare score: 94.28

### #13 ai
- URL: https://www.theagenticleaderboard.com/agent/ai
- Developer: vercel
- Category: Agent Framework
- Repository: https://github.com/vercel/ai
- Overall score: 86.1 / 100
- Description: The AI Toolkit for TypeScript. From the creators of Next.js, the AI SDK is a free open-source library for building AI-powered applications and agents 
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 26113
- Community mindshare score: 85.32

### #14 DeepSeek-Reasonix
- URL: https://www.theagenticleaderboard.com/agent/deepseek-reasonix
- Developer: esengine
- Category: Coding & Software Engineering
- Repository: https://github.com/esengine/DeepSeek-Reasonix
- Overall score: 85.9 / 100
- Description: DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running.
- Reliability success rate: 97.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 33616
- Community mindshare score: 84.79

### #15 jan
- URL: https://www.theagenticleaderboard.com/agent/jan
- Developer: janhq
- Category: Chatbot & Conversational
- Repository: https://github.com/janhq/jan
- Overall score: 85.8 / 100
- Description: Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.
- Reliability success rate: 97.4%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 43944
- Community mindshare score: 86.29

### #16 Agent-Reach
- URL: https://www.theagenticleaderboard.com/agent/agent-reach
- Developer: Panniantong
- Category: Browser & Computer Use
- Repository: https://github.com/Panniantong/Agent-Reach
- Overall score: 85.7 / 100
- Description: Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 70046
- Community mindshare score: 88.87

### #17 career-ops
- URL: https://www.theagenticleaderboard.com/agent/career-ops
- Developer: santifer
- Category: Autonomous Agent
- Repository: https://github.com/santifer/career-ops
- Overall score: 85.6 / 100
- Description: Open-source AI job search: scan job portals, evaluate listings with a structured A-F rubric into a 1.0-5.0 score, tailor your CV, track applications — runs locally in your AI coding CLI (Claude Code, Codex, OpenCode, Antigravity…)
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 63367
- Community mindshare score: 90.17

### #18 llama_index
- URL: https://www.theagenticleaderboard.com/agent/llama-index
- Developer: run-llama
- Category: Agent Framework
- Repository: https://github.com/run-llama/llama_index
- Overall score: 85.5 / 100
- Description: LlamaIndex is the leading document agent and OCR platform
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 51528
- Community mindshare score: 88.61

### #19 last30days-skill
- URL: https://www.theagenticleaderboard.com/agent/last30days-skill
- Developer: mvanhorn
- Category: Research & Reasoning
- Repository: https://github.com/mvanhorn/last30days-skill
- Overall score: 85.5 / 100
- Description: AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
- Reliability success rate: 94.2%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 57794
- Community mindshare score: 88.22

### #20 smolagents
- URL: https://www.theagenticleaderboard.com/agent/smolagents
- Developer: huggingface
- Category: Multi-Modal Agent
- Repository: https://github.com/huggingface/smolagents
- Overall score: 85.5 / 100
- Description: 🤗 smolagents: a barebones library for agents that think in code.
- Reliability success rate: 97.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 28739
- Community mindshare score: 82.58

### #21 cal.diy
- URL: https://www.theagenticleaderboard.com/agent/cal-diy
- Developer: calcom
- Category: Workflow Orchestration
- Repository: https://github.com/calcom/cal.diy
- Overall score: 85.4 / 100
- Description: Scheduling infrastructure for absolutely everyone.
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 47396
- Community mindshare score: 89.29

### #22 WrenAI
- URL: https://www.theagenticleaderboard.com/agent/wrenai
- Developer: Canner
- Category: Data & Analytics
- Repository: https://github.com/Canner/WrenAI
- Overall score: 85.4 / 100
- Description: GenBI (Generative BI) for AI agents, an open-source, governed text-to-SQL through an open context layer that turns natural-language questions into trusted dashboards, charts, and SQL across 20+ data sources, such as BigQuery, Snowflake, PostgreSQL, ClickHouse, Amazon Redshift, Databricks and more.
- Reliability success rate: 97.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 17209
- Community mindshare score: 82.17

### #23 OpenMontage
- URL: https://www.theagenticleaderboard.com/agent/openmontage
- Developer: calesthio
- Category: Multi-Modal Agent
- Repository: https://github.com/calesthio/OpenMontage
- Overall score: 85.3 / 100
- Description: World's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent skill and production-knowledge files. Turn your AI coding assistant into a full video production studio.
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 46524
- Community mindshare score: 86.91

### #24 agents
- URL: https://www.theagenticleaderboard.com/agent/agents
- Developer: wshobson
- Category: Coding & Software Engineering
- Repository: https://github.com/wshobson/agents
- Overall score: 85.3 / 100
- Description: Multi-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLI
- Reliability success rate: 97.0%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 17.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 38673
- Community mindshare score: 85.86

### #25 hermes-agent
- URL: https://www.theagenticleaderboard.com/agent/hermes-agent
- Developer: NousResearch
- Category: General AI Agent
- Repository: https://github.com/NousResearch/hermes-agent
- Overall score: 85.2 / 100
- Description: The agent that grows with you
- Reliability success rate: 95.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 228167
- Community mindshare score: 97.05

### #26 learn-claude-code
- URL: https://www.theagenticleaderboard.com/agent/learn-claude-code
- Developer: shareAI-lab
- Category: Coding & Software Engineering
- Repository: https://github.com/shareAI-lab/learn-claude-code
- Overall score: 85.2 / 100
- Description: Bash is all you need -  A nano claude code–like 「agent harness」, built from 0 to 1
- Reliability success rate: 92.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 73733
- Community mindshare score: 89.3

### #27 deer-flow
- URL: https://www.theagenticleaderboard.com/agent/deer-flow
- Developer: bytedance
- Category: Agent Framework
- Repository: https://github.com/bytedance/deer-flow
- Overall score: 85.1 / 100
- Description: An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.
- Reliability success rate: 97.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 79653
- Community mindshare score: 90.75

### #28 crewAI
- URL: https://www.theagenticleaderboard.com/agent/crewai
- Developer: crewAIInc
- Category: Agent Framework
- Repository: https://github.com/crewAIInc/crewAI
- Overall score: 85.1 / 100
- Description: Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
- Reliability success rate: 92.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 56892
- Community mindshare score: 89.01

### #29 OmniRoute
- URL: https://www.theagenticleaderboard.com/agent/omniroute
- Developer: diegosouzapw
- Category: DevOps & Infrastructure
- Repository: https://github.com/diegosouzapw/OmniRoute
- Overall score: 85.1 / 100
- Description: Never stop coding. Free MIT AI gateway: one endpoint, 290+ providers (90+ free), 500+ models — Kimi, Claude, GPT, OpenAI, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A,
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 44700
- Community mindshare score: 87.61

### #30 lobehub
- URL: https://www.theagenticleaderboard.com/agent/lobehub
- Developer: lobehub
- Category: Workflow Orchestration
- Repository: https://github.com/lobehub/lobehub
- Overall score: 85.0 / 100
- Description: 🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 81461
- Community mindshare score: 91.49

### #31 openclaude
- URL: https://www.theagenticleaderboard.com/agent/openclaude
- Developer: Gitlawb
- Category: General AI Agent
- Repository: https://github.com/Gitlawb/openclaude
- Overall score: 85.0 / 100
- Description: runs anywhere. uses anything
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 30601
- Community mindshare score: 86.94

### #32 ToolJet
- URL: https://www.theagenticleaderboard.com/agent/tooljet
- Developer: ToolJet
- Category: Coding & Software Engineering
- Repository: https://github.com/ToolJet/ToolJet
- Overall score: 84.9 / 100
- Description: ToolJet is the open-source foundation of ToolJet AI - the enterprise app generation platform for building internal tools, dashboard, business applications, workflows and AI agents 🚀
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 38312
- Community mindshare score: 86.8

### #33 semantic-kernel
- URL: https://www.theagenticleaderboard.com/agent/semantic-kernel
- Developer: microsoft
- Category: Agent Framework
- Repository: https://github.com/microsoft/semantic-kernel
- Overall score: 84.9 / 100
- Description: Integrate cutting-edge LLM technology quickly and easily into your apps
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 28437
- Community mindshare score: 85.32

### #34 sim
- URL: https://www.theagenticleaderboard.com/agent/sim
- Developer: simstudioai
- Category: Agent Framework
- Repository: https://github.com/simstudioai/sim
- Overall score: 84.9 / 100
- Description: Build, deploy, and orchestrate AI agents. Sim is the central intelligence layer for your AI workforce.
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 29392
- Community mindshare score: 85.27

### #35 crush
- URL: https://www.theagenticleaderboard.com/agent/crush
- Developer: charmbracelet
- Category: Coding & Software Engineering
- Repository: https://github.com/charmbracelet/crush
- Overall score: 84.9 / 100
- Description: Glamourous agentic coding for all 💘
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 27240
- Community mindshare score: 83.99

### #36 langgraph
- URL: https://www.theagenticleaderboard.com/agent/langgraph
- Developer: langchain-ai
- Category: Agent Framework
- Repository: https://github.com/langchain-ai/langgraph
- Overall score: 84.8 / 100
- Description: Build resilient agents.
- Reliability success rate: 100.0%
- Tool selection quality: 49.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 39356
- Community mindshare score: 87.21

### #37 posthog
- URL: https://www.theagenticleaderboard.com/agent/posthog
- Developer: PostHog
- Category: Data & Analytics
- Repository: https://github.com/PostHog/posthog
- Overall score: 84.8 / 100
- Description: :hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Stee
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 37596
- Community mindshare score: 85.85

### #38 dspy
- URL: https://www.theagenticleaderboard.com/agent/dspy
- Developer: stanfordnlp
- Category: Agent Framework
- Repository: https://github.com/stanfordnlp/dspy
- Overall score: 84.8 / 100
- Description: DSPy: The framework for programming—not prompting—language models
- Reliability success rate: 95.0%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 36990
- Community mindshare score: 85.59

### #39 OpenCLI
- URL: https://www.theagenticleaderboard.com/agent/opencli
- Developer: jackwener
- Category: Browser & Computer Use
- Repository: https://github.com/jackwener/OpenCLI
- Overall score: 84.8 / 100
- Description: Make Any Website into CLI & Use your logged-in browser by AI agent. 
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 28005
- Community mindshare score: 84.54

### #40 deepagents
- URL: https://www.theagenticleaderboard.com/agent/deepagents
- Developer: langchain-ai
- Category: Agent Framework
- Repository: https://github.com/langchain-ai/deepagents
- Overall score: 84.7 / 100
- Description: The batteries-included agent harness.
- Reliability success rate: 100.0%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 27592
- Community mindshare score: 84.96

### #41 kestra
- URL: https://www.theagenticleaderboard.com/agent/kestra
- Developer: kestra-io
- Category: Workflow Orchestration
- Repository: https://github.com/kestra-io/kestra
- Overall score: 84.7 / 100
- Description: Event Driven Orchestration & Scheduling Platform for Mission Critical Applications
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 27704
- Community mindshare score: 84.58

### #42 cc-haha
- URL: https://www.theagenticleaderboard.com/agent/cc-haha
- Developer: NanmiCoder
- Category: Coding & Software Engineering
- Repository: https://github.com/NanmiCoder/cc-haha
- Overall score: 84.7 / 100
- Description: Local-first cross-platform desktop workspace for Claude Code / agents: multi-agent, Git worktrees, code diffs, skill marketplace, multi-model, Computer Use, task-aware desktop pets, with WeChat, Feishu, DingTalk, Telegram, WhatsApp and H5 access.
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 14021
- Community mindshare score: 84.04

### #43 pydantic-ai
- URL: https://www.theagenticleaderboard.com/agent/pydantic-ai
- Developer: pydantic
- Category: Agent Framework
- Repository: https://github.com/pydantic/pydantic-ai
- Overall score: 84.7 / 100
- Description: AI Agent Framework, the Pydantic way
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 19187
- Community mindshare score: 83.0

### #44 cognee
- URL: https://www.theagenticleaderboard.com/agent/cognee
- Developer: topoteretes
- Category: Data & Analytics
- Repository: https://github.com/topoteretes/cognee
- Overall score: 84.6 / 100
- Description: Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 29907
- Community mindshare score: 84.87

### #45 budibase
- URL: https://www.theagenticleaderboard.com/agent/budibase
- Developer: Budibase
- Category: Workflow Orchestration
- Repository: https://github.com/Budibase/budibase
- Overall score: 84.6 / 100
- Description: AI agents, automations and apps that run your operations. Model agnostic.
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 28191
- Community mindshare score: 84.16

### #46 ruflo
- URL: https://www.theagenticleaderboard.com/agent/ruflo
- Developer: ruvnet
- Category: Agent Framework
- Repository: https://github.com/ruvnet/ruflo
- Overall score: 84.5 / 100
- Description: 🌊 The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
- Reliability success rate: 97.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 67538
- Community mindshare score: 89.62

### #47 anything-llm
- URL: https://www.theagenticleaderboard.com/agent/anything-llm
- Developer: Mintplex-Labs
- Category: Chatbot & Conversational
- Repository: https://github.com/Mintplex-Labs/anything-llm
- Overall score: 84.5 / 100
- Description: Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience 
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 64547
- Community mindshare score: 89.01

### #48 signoz
- URL: https://www.theagenticleaderboard.com/agent/signoz
- Developer: SigNoz
- Category: DevOps & Infrastructure
- Repository: https://github.com/SigNoz/signoz
- Overall score: 84.5 / 100
- Description: SigNoz is an open-source, OpenTelemetry-native observability platform for your team and their AI agents. Get logs, metrics, and traces in one tool with features like APM, distributed tracing, log management, infra monitoring, etc. Combined with SigNoz MCP and a native AI teammate (in SigNoz Cloud) i
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 31812
- Community mindshare score: 84.77

### #49 headroom
- URL: https://www.theagenticleaderboard.com/agent/headroom
- Developer: headroomlabs-ai
- Category: DevOps & Infrastructure
- Repository: https://github.com/headroomlabs-ai/headroom
- Overall score: 84.4 / 100
- Description: Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 65731
- Community mindshare score: 88.68

### #50 ai-engineering-from-scratch
- URL: https://www.theagenticleaderboard.com/agent/ai-engineering-from-scratch
- Developer: rohitg00
- Category: General AI Agent
- Repository: https://github.com/rohitg00/ai-engineering-from-scratch
- Overall score: 84.4 / 100
- Description: Learn it. Build it. Ship it for others.
- Reliability success rate: 97.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 46420
- Community mindshare score: 88.27

### #51 gemini-cli
- URL: https://www.theagenticleaderboard.com/agent/gemini-cli
- Developer: google-gemini
- Category: General AI Agent
- Repository: https://github.com/google-gemini/gemini-cli
- Overall score: 84.3 / 100
- Description: An open-source AI agent that brings the power of Gemini directly into your terminal.
- Reliability success rate: 95.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 106435
- Community mindshare score: 92.29

### #52 agent-skills
- URL: https://www.theagenticleaderboard.com/agent/agent-skills
- Developer: addyosmani
- Category: Coding & Software Engineering
- Repository: https://github.com/addyosmani/agent-skills
- Overall score: 84.3 / 100
- Description: Production-grade engineering skills for AI coding agents.
- Reliability success rate: 91.4%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 17.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 85435
- Community mindshare score: 90.59

### #53 nanoclaw
- URL: https://www.theagenticleaderboard.com/agent/nanoclaw
- Developer: nanocoai
- Category: Chatbot & Conversational
- Repository: https://github.com/nanocoai/nanoclaw
- Overall score: 84.2 / 100
- Description: A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 30478
- Community mindshare score: 87.57

### #54 daily_stock_analysis
- URL: https://www.theagenticleaderboard.com/agent/daily-stock-analysis
- Developer: ZhuLinsen
- Category: Data & Analytics
- Repository: https://github.com/ZhuLinsen/daily_stock_analysis
- Overall score: 84.1 / 100
- Description: LLM 驱动的多市场股票智能分析系统：多源行情、实时新闻、决策看板与自动推送，支持零成本定时运行。  LLM-powered multi-market stock analysis system with multi-source market data, real-time news, decision dashboard, automated notifications, and cost-free scheduled runs.
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 61649
- Community mindshare score: 92.59

### #55 goose
- URL: https://www.theagenticleaderboard.com/agent/goose
- Developer: aaif-goose
- Category: Coding & Software Engineering
- Repository: https://github.com/aaif-goose/goose
- Overall score: 84.1 / 100
- Description: an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
- Reliability success rate: 97.4%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 52623
- Community mindshare score: 88.19

### #56 tidb
- URL: https://www.theagenticleaderboard.com/agent/tidb
- Developer: pingcap
- Category: Data & Analytics
- Repository: https://github.com/pingcap/tidb
- Overall score: 84.1 / 100
- Description: TiDB is built for agentic workloads that grow unpredictably, with ACID guarantees and native support for transactions, analytics, and vector search. No data silos. No noisy neighbors. No infrastructure ceiling.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 40412
- Community mindshare score: 87.31

### #57 Vibe-Trading
- URL: https://www.theagenticleaderboard.com/agent/vibe-trading
- Developer: HKUDS
- Category: Autonomous Agent
- Repository: https://github.com/HKUDS/Vibe-Trading
- Overall score: 84.1 / 100
- Description: "Vibe-Trading: Your Personal Trading Agent"
- Reliability success rate: 97.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 30510
- Community mindshare score: 85.88

### #58 gpt-researcher
- URL: https://www.theagenticleaderboard.com/agent/gpt-researcher
- Developer: assafelovic
- Category: Research & Reasoning
- Repository: https://github.com/assafelovic/gpt-researcher
- Overall score: 84.1 / 100
- Description: An autonomous agent that conducts deep research on any data using any LLM providers
- Reliability success rate: 100.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 28910
- Community mindshare score: 82.72

### #59 ollama
- URL: https://www.theagenticleaderboard.com/agent/ollama
- Developer: ollama
- Category: DevOps & Infrastructure
- Repository: https://github.com/ollama/ollama
- Overall score: 84.0 / 100
- Description: Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
- Reliability success rate: 92.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 178185
- Community mindshare score: 94.48

### #60 langflow
- URL: https://www.theagenticleaderboard.com/agent/langflow
- Developer: langflow-ai
- Category: Workflow Orchestration
- Repository: https://github.com/langflow-ai/langflow
- Overall score: 84.0 / 100
- Description: Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
- Reliability success rate: 95.0%
- Tool selection quality: 69.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 153012
- Community mindshare score: 92.93

### #61 awesome-copilot
- URL: https://www.theagenticleaderboard.com/agent/awesome-copilot
- Developer: github
- Category: Coding & Software Engineering
- Repository: https://github.com/github/awesome-copilot
- Overall score: 84.0 / 100
- Description: Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 37640
- Community mindshare score: 86.57

### #62 hyperframes
- URL: https://www.theagenticleaderboard.com/agent/hyperframes
- Developer: heygen-com
- Category: Multi-Modal Agent
- Repository: https://github.com/heygen-com/hyperframes
- Overall score: 84.0 / 100
- Description: Write HTML. Render video. Built for agents.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 40326
- Community mindshare score: 86.45

### #63 composio
- URL: https://www.theagenticleaderboard.com/agent/composio
- Developer: ComposioHQ
- Category: Agent Framework
- Repository: https://github.com/ComposioHQ/composio
- Overall score: 84.0 / 100
- Description: Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
- Reliability success rate: 97.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 29612
- Community mindshare score: 85.69

### #64 edict
- URL: https://www.theagenticleaderboard.com/agent/edict
- Developer: cft0808
- Category: Agent Framework
- Repository: https://github.com/cft0808/edict
- Overall score: 84.0 / 100
- Description: 🏛️ 三省六部制 · OpenClaw Multi-Agent Orchestration System — 9 specialized AI agents with real-time dashboard, model config, and full audit trails
- Reliability success rate: 100.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 16345
- Community mindshare score: 81.09

### #65 ponytail
- URL: https://www.theagenticleaderboard.com/agent/ponytail
- Developer: DietrichGebert
- Category: Coding & Software Engineering
- Repository: https://github.com/DietrichGebert/ponytail
- Overall score: 83.9 / 100
- Description: Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
- Reliability success rate: 88.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 99799
- Community mindshare score: 90.13

### #66 mem0
- URL: https://www.theagenticleaderboard.com/agent/mem0
- Developer: mem0ai
- Category: Agent Framework
- Repository: https://github.com/mem0ai/mem0
- Overall score: 83.9 / 100
- Description: Universal memory layer for AI Agents
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 62928
- Community mindshare score: 88.98

### #67 openinterpreter
- URL: https://www.theagenticleaderboard.com/agent/openinterpreter
- Developer: openinterpreter
- Category: Browser & Computer Use
- Repository: https://github.com/openinterpreter/openinterpreter
- Overall score: 83.9 / 100
- Description: A coding agent for open models like Kimi K3
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 67934
- Community mindshare score: 88.96

### #68 zeroclaw
- URL: https://www.theagenticleaderboard.com/agent/zeroclaw
- Developer: zeroclaw-labs
- Category: Agent Framework
- Repository: https://github.com/zeroclaw-labs/zeroclaw
- Overall score: 83.9 / 100
- Description: Fast, small, and fully autonomous AI personal assistant infrastructure, any OS, any platform — deploy anywhere, swap anything 🦀
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 32549
- Community mindshare score: 86.1

### #69 openhuman
- URL: https://www.theagenticleaderboard.com/agent/openhuman
- Developer: tinyhumansai
- Category: General AI Agent
- Repository: https://github.com/tinyhumansai/openhuman
- Overall score: 83.9 / 100
- Description: Your Personal AI super intelligence. A brain that builds a local-first memory of your life, a fantastic orchestrator of agent fleets and workflows, and a deep researcher.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 36142
- Community mindshare score: 85.93

### #70 planning-with-files
- URL: https://www.theagenticleaderboard.com/agent/planning-with-files
- Developer: OthmanAdi
- Category: Coding & Software Engineering
- Repository: https://github.com/OthmanAdi/planning-with-files
- Overall score: 83.9 / 100
- Description: Persistent file-based planning for AI coding agents and long-running tasks. Crash-proof markdown plans, session recovery after /clear and compaction, per-turn re-injection against context rot, deterministic completion gate. Manus-style. Install from npm, the Claude Code plugin marketplace, or npx sk
- Reliability success rate: 94.2%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 17.5
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 26079
- Community mindshare score: 83.76

### #71 UI-TARS-desktop
- URL: https://www.theagenticleaderboard.com/agent/ui-tars-desktop
- Developer: bytedance
- Category: Browser & Computer Use
- Repository: https://github.com/bytedance/UI-TARS-desktop
- Overall score: 83.8 / 100
- Description: The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 38537
- Community mindshare score: 85.75

### #72 firecrawl
- URL: https://www.theagenticleaderboard.com/agent/firecrawl
- Developer: firecrawl
- Category: Browser & Computer Use
- Repository: https://github.com/firecrawl/firecrawl
- Overall score: 83.7 / 100
- Description: The context API to search, scrape, and interact with the web at scale. 🔥
- Reliability success rate: 92.4%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 164516
- Community mindshare score: 93.08

### #73 claude-mem
- URL: https://www.theagenticleaderboard.com/agent/claude-mem
- Developer: thedotmack
- Category: Agent Framework
- Repository: https://github.com/thedotmack/claude-mem
- Overall score: 83.7 / 100
- Description: Persistent Context Across Sessions for Every Agent –  Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 90277
- Community mindshare score: 90.62

### #74 activepieces
- URL: https://www.theagenticleaderboard.com/agent/activepieces
- Developer: activepieces
- Category: Workflow Orchestration
- Repository: https://github.com/activepieces/activepieces
- Overall score: 83.6 / 100
- Description: AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 23695
- Community mindshare score: 84.6

### #75 kilocode
- URL: https://www.theagenticleaderboard.com/agent/kilocode
- Developer: Kilo-Org
- Category: Coding & Software Engineering
- Repository: https://github.com/Kilo-Org/kilocode
- Overall score: 83.6 / 100
- Description: Kilo is the all-in-one agentic engineering platform. Build, ship, and iterate faster with the most popular open source coding agent.
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 26805
- Community mindshare score: 84.55

### #76 FinceptTerminal
- URL: https://www.theagenticleaderboard.com/agent/finceptterminal
- Developer: Fincept-Corporation
- Category: Data & Analytics
- Repository: https://github.com/Fincept-Corporation/FinceptTerminal
- Overall score: 83.6 / 100
- Description: FinceptTerminal is a modern finance application offering advanced market analytics, investment research, and economic data tools, designed for interactive exploration and data-driven decision-making in a user-friendly environment.
- Reliability success rate: 97.4%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 30070
- Community mindshare score: 83.9

### #77 CowAgent
- URL: https://www.theagenticleaderboard.com/agent/cowagent
- Developer: zhayujie
- Category: Chatbot & Conversational
- Repository: https://github.com/zhayujie/CowAgent
- Overall score: 83.5 / 100
- Description: Open-source super AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-model, multi-channel. Lightweight, extensible, one-line install. (formerly chatgpt-on-wechat)
- Reliability success rate: 94.2%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 46442
- Community mindshare score: 88.7

### #78 conductor
- URL: https://www.theagenticleaderboard.com/agent/conductor
- Developer: conductor-oss
- Category: Workflow Orchestration
- Repository: https://github.com/conductor-oss/conductor
- Overall score: 83.5 / 100
- Description: Conductor is an event driven agentic workflow engine providing durable and highly resilient execution engine for applications and AI Agents
- Reliability success rate: 97.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 32072
- Community mindshare score: 82.99

### #79 agency-agents
- URL: https://www.theagenticleaderboard.com/agent/agency-agents
- Developer: msitarzewski
- Category: General AI Agent
- Repository: https://github.com/msitarzewski/agency-agents
- Overall score: 83.4 / 100
- Description: A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
- Reliability success rate: 97.0%
- Tool selection quality: 31.2%
- Autonomous iteration avg steps: 15.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 141320
- Community mindshare score: 93.81

### #80 pi
- URL: https://www.theagenticleaderboard.com/agent/pi
- Developer: earendil-works
- Category: Agent Framework
- Repository: https://github.com/earendil-works/pi
- Overall score: 83.4 / 100
- Description: AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI
- Reliability success rate: 92.4%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 86381
- Community mindshare score: 91.01

### #81 LibreChat
- URL: https://www.theagenticleaderboard.com/agent/librechat
- Developer: danny-avila
- Category: Chatbot & Conversational
- Repository: https://github.com/danny-avila/LibreChat
- Overall score: 83.4 / 100
- Description: Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message search, Code Interpreter, langchain, DALL-E-3, OpenAPI Actions, Functions, Secure Multi-User A
- Reliability success rate: 100.0%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 41875
- Community mindshare score: 88.02

### #82 NemoClaw
- URL: https://www.theagenticleaderboard.com/agent/nemoclaw
- Developer: NVIDIA
- Category: DevOps & Infrastructure
- Repository: https://github.com/NVIDIA/NemoClaw
- Overall score: 83.4 / 100
- Description: Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 22111
- Community mindshare score: 83.84

### #83 ragflow
- URL: https://www.theagenticleaderboard.com/agent/ragflow
- Developer: infiniflow
- Category: Data & Analytics
- Repository: https://github.com/infiniflow/ragflow
- Overall score: 83.3 / 100
- Description: RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
- Reliability success rate: 92.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 87173
- Community mindshare score: 90.97

### #84 open-design
- URL: https://www.theagenticleaderboard.com/agent/open-design
- Developer: nexu-io
- Category: Coding & Software Engineering
- Repository: https://github.com/nexu-io/open-design
- Overall score: 83.3 / 100
- Description: 🎨 The open-source Claude Design alternative. 🖥️ Local-first desktop app. 🖼️ Your coding agent becomes the design engine: prototypes, landing pages, dashboards, slides, images & video — real files, HTML/PDF/PPTX/MP4 export. 🤖 Claude Code / Codex / Cursor / Gemini / OpenCode / Qwen & 20+ CLIs via BYOK
- Reliability success rate: 92.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 84827
- Community mindshare score: 90.81

### #85 oh-my-openagent
- URL: https://www.theagenticleaderboard.com/agent/oh-my-openagent
- Developer: code-yeongyu
- Category: Coding & Software Engineering
- Repository: https://github.com/code-yeongyu/oh-my-openagent
- Overall score: 83.3 / 100
- Description: omo/lazycodex: The coding agent for tokenmaxxers;the one and only agent harness for complex codebases. For your Codex, for your OpenCode
- Reliability success rate: 94.2%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 67591
- Community mindshare score: 88.95

### #86 qlib
- URL: https://www.theagenticleaderboard.com/agent/qlib
- Developer: microsoft
- Category: Data & Analytics
- Repository: https://github.com/microsoft/qlib
- Overall score: 83.3 / 100
- Description: Qlib is an AI-oriented Quant investment platform that aims to use AI tech to empower Quant Research, from exploring ideas to implementing productions. Qlib supports diverse ML modeling paradigms, including supervised learning, market dynamics modeling, and RL, and is now equipped with https://github
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 47240
- Community mindshare score: 86.2

### #87 adk-python
- URL: https://www.theagenticleaderboard.com/agent/adk-python
- Developer: google
- Category: Agent Framework
- Repository: https://github.com/google/adk-python
- Overall score: 83.3 / 100
- Description: An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
- Reliability success rate: 97.4%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 21062
- Community mindshare score: 84.09

### #88 agentmemory
- URL: https://www.theagenticleaderboard.com/agent/agentmemory
- Developer: rohitg00
- Category: Coding & Software Engineering
- Repository: https://github.com/rohitg00/agentmemory
- Overall score: 83.3 / 100
- Description: #1 Persistent memory for AI coding agents based on real-world benchmarks
- Reliability success rate: 97.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 26821
- Community mindshare score: 84.05

### #89 jcode
- URL: https://www.theagenticleaderboard.com/agent/jcode
- Developer: 1jehuang
- Category: Coding & Software Engineering
- Repository: https://github.com/1jehuang/jcode
- Overall score: 83.3 / 100
- Description: The most RAM efficient harness
- Reliability success rate: 91.4%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.02
- Latency per step: 500.0ms
- GitHub stars: 16685
- Community mindshare score: 82.0

### #90 continue
- URL: https://www.theagenticleaderboard.com/agent/continue
- Developer: continuedev
- Category: Coding & Software Engineering
- Repository: https://github.com/continuedev/continue
- Overall score: 83.2 / 100
- Description: open-source coding agent
- Reliability success rate: 97.4%
- Tool selection quality: 67.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 35423
- Community mindshare score: 86.52

### #91 mlflow
- URL: https://www.theagenticleaderboard.com/agent/mlflow
- Developer: mlflow
- Category: DevOps & Infrastructure
- Repository: https://github.com/mlflow/mlflow
- Overall score: 83.2 / 100
- Description: The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data.
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 27446
- Community mindshare score: 85.88

### #92 openai-agents-python
- URL: https://www.theagenticleaderboard.com/agent/openai-agents-python
- Developer: openai
- Category: Agent Framework
- Repository: https://github.com/openai/openai-agents-python
- Overall score: 83.2 / 100
- Description: A lightweight, powerful framework for multi-agent workflows
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 28533
- Community mindshare score: 85.47

### #93 oh-my-pi
- URL: https://www.theagenticleaderboard.com/agent/oh-my-pi
- Developer: can1357
- Category: Coding & Software Engineering
- Repository: https://github.com/can1357/oh-my-pi
- Overall score: 83.2 / 100
- Description: ⌥  AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more
- Reliability success rate: 97.0%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 23455
- Community mindshare score: 83.52

### #94 hive
- URL: https://www.theagenticleaderboard.com/agent/hive
- Developer: aden-hive
- Category: Agent Framework
- Repository: https://github.com/aden-hive/hive
- Overall score: 83.2 / 100
- Description: Multi-Agent Harness for Production AI
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 10890
- Community mindshare score: 82.4

### #95 OpenHands
- URL: https://www.theagenticleaderboard.com/agent/openhands
- Developer: OpenHands
- Category: Coding & Software Engineering
- Repository: https://github.com/OpenHands/OpenHands
- Overall score: 83.1 / 100
- Description: 🙌 OpenHands: AI-Driven Development
- Reliability success rate: 92.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 83603
- Community mindshare score: 90.91

### #96 agno
- URL: https://www.theagenticleaderboard.com/agent/agno
- Developer: agno-agi
- Category: Agent Framework
- Repository: https://github.com/agno-agi/agno
- Overall score: 83.1 / 100
- Description: Build, run, and manage agent platforms.
- Reliability success rate: 95.0%
- Tool selection quality: 59.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 41646
- Community mindshare score: 87.28

### #97 agentscope
- URL: https://www.theagenticleaderboard.com/agent/agentscope
- Developer: agentscope-ai
- Category: Agent Framework
- Repository: https://github.com/agentscope-ai/agentscope
- Overall score: 83.1 / 100
- Description: Build and run agents you can see, understand and trust.
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 28766
- Community mindshare score: 84.97

### #98 SWE-agent
- URL: https://www.theagenticleaderboard.com/agent/swe-agent
- Developer: SWE-agent
- Category: Coding & Software Engineering
- Repository: https://github.com/SWE-agent/SWE-agent
- Overall score: 83.1 / 100
- Description: SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024] 
- Reliability success rate: 97.4%
- Tool selection quality: 71.2%
- Autonomous iteration avg steps: 25.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 20037
- Community mindshare score: 82.26

### #99 agent-framework
- URL: https://www.theagenticleaderboard.com/agent/agent-framework
- Developer: microsoft
- Category: Agent Framework
- Repository: https://github.com/microsoft/agent-framework
- Overall score: 83.1 / 100
- Description: A framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.
- Reliability success rate: 97.4%
- Tool selection quality: 51.2%
- Autonomous iteration avg steps: 20.0
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 12707
- Community mindshare score: 81.24

### #100 OpenViking
- URL: https://www.theagenticleaderboard.com/agent/openviking
- Developer: volcengine
- Category: Data & Analytics
- Repository: https://github.com/volcengine/OpenViking
- Overall score: 82.9 / 100
- Description: Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
- Reliability success rate: 95.0%
- Tool selection quality: 61.2%
- Autonomous iteration avg steps: 22.5
- Cost per task: $0.05
- Latency per step: 500.0ms
- GitHub stars: 28152
- Community mindshare score: 84.18