Report Date: 2026-03-02
Analysis Window: 30 days
Languages Tracked: python, typescript, rust, go, javascript
Top N: 50
Overview
根据数据显示，近期GitHub AI开源项目的趋势高度集中在“AI Agent”领域。Agent的应用、基础设施和生态工具是当前最核心的增长驱动力。同时，为了追求性能 与安全，使用Rust、Go等系统级编程语言构建AI基础设施的趋势也日益明显。
Top 50 Repositories
| Rank | Repository | Category | Language | Heat_7 | Heat_30 | Score |
|---|---|---|---|---|---|---|
| 1 | openclaw | typescript | 0 | 106,934 | 1,996 | |
| 2 | shannon | agent | typescript | 0 | 21,789 | 72 |
| 3 | claude-mem | agent | typescript | 0 | 16,380 | 54 |
| 4 | pi-mono | agent | typescript | 3,773 | 13,950 | 208 |
| 5 | langextract | llm | python | 0 | 10,338 | 34 |
| 6 | PageIndex | rag | python | 3,722 | 8,144 | 186 |
| 7 | dexter | agent | typescript | 0 | 7,344 | 24 |
| 8 | memU | agent | python | 0 | 5,200 | 211 |
| 9 | codex | agent | rust | 0 | 4,508 | 15 |
| 10 | chrome-devtools-mcp | agent | typescript | 0 | 4,254 | 14 |
| 11 | AstrBot | llm | python | 0 | 3,078 | 10 |
| 12 | litellm | llm | python | 0 | 2,500 | 8 |
| 13 | TradingAgents | agent | python | 0 | 2,263 | 7 |
| 14 | ruvector | rust | 1,776 | 1,961 | 82 | |
| 15 | superset | agent | typescript | 0 | 1,612 | 238 |
| 16 | agents | agent | typescript | 776 | 1,350 | 37 |
| 17 | meetily | rust | 0 | 550 | 1 | |
| 18 | agent-deck | llm | go | 0 | 538 | 1 |
| 19 | screenpipe | rust | 0 | 480 | 1 | |
| 20 | rig | llm | rust | 111 | 448 | 6 |
| 21 | skills | agent | javascript | 0 | 429 | 1 |
| 22 | pi-skills | agent | javascript | 124 | 403 | 6 |
| 23 | google_workspace_mcp | python | 0 | 341 | 1 | |
| 24 | atlassian-mcp-server | agent | javascript | 21 | 91 | 1 |
| 25 | OpenSandbox | agent | python | 2,012 | 0 | 793 |
| 26 | skills | agent | python | 6,949 | 0 | 734 |
| 27 | ruflo | agent | typescript | 2,786 | 0 | 579 |
| 28 | GitNexus | agent | typescript | 5,791 | 0 | 248 |
| 29 | Agent-Skills-for-Context-Engineering | agent | python | 4,368 | 0 | 187 |
| 30 | MobileAgent | agent | python | 408 | 0 | 131 |
| 31 | ai-guide | llm | javascript | 1,442 | 0 | 131 |
| 32 | claude-scientific-skills | agent | python | 0 | 0 | 113 |
| 33 | pentagi | agent | go | 2,280 | 0 | 97 |
| 34 | agents | agent | python | 0 | 0 | 94 |
| 35 | beads | agent | go | 0 | 0 | 81 |
| 36 | agent-skills | agent | javascript | 0 | 0 | 78 |
| 37 | Vision-Agents | agent | python | 0 | 0 | 72 |
| 38 | OpenBB | agent | python | 1,577 | 0 | 67 |
| 39 | claudian | llm | typescript | 0 | 0 | 38 |
| 40 | crewAI | agent | python | 0 | 0 | 37 |
| 41 | cherry-studio | agent | typescript | 0 | 0 | 30 |
| 42 | plano | agent | rust | 685 | 0 | 29 |
| 43 | manifest | llm | typescript | 0 | 0 | 27 |
| 44 | lobehub | agent | typescript | 0 | 0 | 25 |
| 45 | keploy | agent | go | 468 | 0 | 20 |
| 46 | monty | rust | 0 | 0 | 19 | |
| 47 | opik | llm | python | 0 | 0 | 18 |
| 48 | servo | rag | rust | 0 | 0 | 17 |
| 49 | anything-llm | agent | javascript | 405 | 0 | 17 |
| 50 | nofx | go | 219 | 0 | 9 |
Category Breakdown
(7 projects)
未分类项目占据一定数量，表明AI领域的新概念和交叉方向仍在不断涌现，探索新的可能性。
Average Heat_7: 285
Average Score: 301
- openclaw - Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
- ruvector - RuVector is a High Performance, Real-Time, Self-Learning, Vector Graph Neural Network, and Database built in Rust.
- meetily - Privacy first, AI meeting assistant with 4x faster Parakeet/Whisper live transcription, speaker diarization, and Ollama summarization built on Rust. 100% local processing. no cloud required. Meetil...
- screenpipe - screenpipe turns your computer into a personal AI that knows everything you've done. record. search. automate. all local, all private, all yours.
- google_workspace_mcp - Control Gmail, Google Calendar, Docs, Sheets, Slides, Chat, Forms, Tasks, Search & Drive with AI - Comprehensive Google Workspace / G Suite MCP Server & CLI Tool
- monty - A minimal, secure Python interpreter written in Rust for use by AI
- nofx - AI Trading OS: Multi-AI, multi-exchange trading infrastructure with Strategy Studio.
agent (32 projects)
Agent是绝对的主导类别，项目数量（32个）和平均热度（1013）均遥遥领先，覆盖编码、金融、安全等多个场景，显示出其广泛的应用探索和落地潜力。
Average Heat_7: 1,013
Average Score: 135
- shannon - Fully autonomous AI hacker to find actual exploits in your web apps. Shannon has achieved a 96.15% success rate on the hint-free, source-aware XBOW Benchmark.
- claude-mem - A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future ...
- pi-mono - AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods
- dexter - An autonomous agent for deep financial research
- memU - Memory for 24/7 proactive agents like openclaw (moltbot, clawdbot).
- codex - Lightweight coding agent that runs in your terminal
- chrome-devtools-mcp - Chrome DevTools for coding agents
- TradingAgents - TradingAgents: Multi-Agents LLM Financial Trading Framework
- superset - IDE for the AI Agents Era - Run an army of Claude Code, Codex, etc. on your machine
- agents - Build and deploy AI Agents on Cloudflare
- skills - A collection of AI agent skills for working with Expo projects and Expo Application Services
- pi-skills - Skills for pi coding agent (compatible with Claude Code and Codex CLI)
- atlassian-mcp-server - Remote MCP Server that securely connects Jira and Confluence with your LLM, IDE, or agent platform of choice.
- OpenSandbox - OpenSandbox is a general-purpose sandbox platform for AI applications, offering multi-language SDKs, unified sandbox APIs, and Docker/Kubernetes runtimes for scenarios like Coding Agents, GUI Agent...
- skills - Public repository for Agent Skills
- ruflo - 🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade arch...
- GitNexus - GitNexus: The Zero-Server Code Intelligence Engine - GitNexus is a client-side knowledge graph creator that runs entirely in your browser. Drop in a GitHub repo or ZIP file, and get an interactive ...
- Agent-Skills-for-Context-Engineering - A comprehensive collection of Agent Skills for context engineering, multi-agent architectures, and production agent systems. Use when building, optimizing, or debugging agent systems that require e...
- MobileAgent - Mobile-Agent: The Powerful GUI Agent Family
- claude-scientific-skills - A set of ready to use Agent Skills for research, science, engineering, analysis, finance and writing.
- pentagi - ✨ Fully autonomous AI Agents system capable of performing complex penetration testing tasks
- agents - Intelligent automation and multi-agent orchestration for Claude Code
- beads - Beads - A memory upgrade for your coding agent
- agent-skills - Vercel's official collection of agent skills
- Vision-Agents - Open Vision Agents by Stream. Build Vision Agents quickly with any model or video provider. Uses Stream's edge network for ultra-low latency.
- OpenBB - Financial data platform for analysts, quants and AI agents.
- crewAI - Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
- cherry-studio - AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs
- plano - Plano is an AI-native proxy server and data plane for agentic apps - centralizing orchestration, safety, observability, and smart LLM routing so you can ship agents faster.
- lobehub - The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboratio...
- keploy - API, Integration, E2E Testing Agent for Developers that actually work. Generate tests, mocks/stubs for your APIs!
- anything-llm - The all-in-one Desktop & Docker AI application with built-in RAG, AI agents, No-code agent builder, MCP compatibility, and more.
llm (9 projects)
LLM相关项目热度相对平稳，趋势更多地指向对大模型的应用层封装、代理和工具化，而非基础模型研究本身。
Average Heat_7: 172
Average Score: 30
- langextract - A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
- AstrBot - Agentic IM Chatbot infrastructure that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨
- litellm - Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthr...
- agent-deck - Terminal session manager for AI coding agents. One TUI for Claude, Gemini, OpenCode, Codex, and more.
- rig - ⚙️🦀 Build modular and scalable LLM Applications in Rust
- ai-guide - 程序员鱼皮的 AI 资源大全 + Vibe Coding 零基础教程，分享大模型选择指南（DeepSeek / GPT / Gemini / Claude）、最新 AI 资讯、Prompt 提示词大全、AI 知识百科（...
- claudian - An Obsidian plugin that embeds Claude Code as an AI collaborator in your vault
- manifest - Smart LLM routing for OpenClaw. Cut Costs up to 70%
- opik - Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.
rag (2 projects)
RAG类别虽然项目不多（2个），但平均热度（1861）位居榜首，这表明该领域正涌现出备受关注的关键技术创新，例如“无向量RAG”。
Average Heat_7: 1,861
Average Score: 101
- PageIndex - 📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG
- servo - Servo aims to empower developers with a lightweight, high-performance alternative for embedding web technologies in applications.
Dark Horse Projects
“黑马”项目（得分高的新项目）普遍展现出近7日内热度的急剧攀升。例如，alibaba/OpenSandbox 和 anthropics/skills 的热度几乎完全集中在最近一周，这表明拥有大厂背景或与热门模型（如Claude）强相关的项目能够迅速获得社区的高度关注。Agent的安全执行环境（沙箱）和能力扩展（Skills）是当前开发者社区关注的焦点。
| Repository | Score | Heat_30 | Heat_7 | Category |
|---|---|---|---|---|
| openclaw | 1,996 | 106,934 | 0 | |
| pi-mono | 208 | 13,950 | 3,773 | agent |
| PageIndex | 186 | 8,144 | 3,722 | rag |
| memU | 211 | 5,200 | 0 | agent |
| superset | 238 | 1,612 | 0 | agent |
| OpenSandbox | 793 | 0 | 2,012 | agent |
| skills | 734 | 0 | 6,949 | agent |
| ruflo | 579 | 0 | 2,786 | agent |
| GitNexus | 248 | 0 | 5,791 | agent |
| Agent-Skills-for-Context-Engineering | 187 | 0 | 4,368 | agent |
| MobileAgent | 131 | 0 | 408 | agent |
| ai-guide | 131 | 0 | 1,442 | llm |
| claude-scientific-skills | 113 | 0 | 0 | agent |
Consecutive Appearances
多个项目（如 openclaw/openclaw, anthropics/skills）连续数周上榜，证明它们不仅拥有初期的爆发热度，也具备吸引社区持续关注的长期价值。部分项目近期热度下降（如 openclaw），可能意味着其发展已从爆发期过渡到稳定期；而另一些项目（如 anthropics/skills）则热度不减，显示其发展势头仍在加速。
| Repository | Weeks in Top | Category | Current Heat_7 |
|---|---|---|---|
| openclaw | 6 | 0 | |
| shannon | 6 | agent | 0 |
| claude-mem | 6 | agent | 0 |
| pi-mono | 6 | agent | 3,773 |
| langextract | 4 | llm | 0 |
| PageIndex | 6 | rag | 3,722 |
| dexter | 6 | agent | 0 |
| memU | 6 | agent | 0 |
| codex | 6 | agent | 0 |
| chrome-devtools-mcp | 3 | agent | 0 |
| AstrBot | 2 | llm | 0 |
| litellm | 2 | llm | 0 |
| superset | 2 | agent | 0 |
| agents | 2 | agent | 776 |
| agent-deck | 2 | llm | 0 |
| screenpipe | 3 | 0 | |
| rig | 2 | llm | 111 |
| skills | 2 | agent | 0 |
| skills | 6 | agent | 6,949 |
| GitNexus | 2 | agent | 5,791 |
| pentagi | 2 | agent | 2,280 |
| OpenBB | 2 | agent | 1,577 |
| lobehub | 2 | agent | 0 |
| keploy | 2 | agent | 468 |
| nofx | 2 | 219 |
Highlighted Repositories
alibaba/OpenSandbox
该项目由阿里巴巴推出，专注于为AI应用提供通用的沙箱平台。其高分和近期爆发的热度反映出市场对于安全、可控地执行AI Agent代码的强烈需求。随着Agent能力增强，如何管理其执行风险已成为一个关键的基础设施问题，而OpenSandbox正为此提供了一个解决方案。
VectifyAI/PageIndex
此项目是RAG领域的显著技术创新，提出了“无向量、基于推理”的新范式。在传统RAG严重依赖向量搜索的背景下，这种新方法获得了社区的高度关注。其高热度表明，开发者正在积极探索超越纯向量检索、能支持更复杂逻辑推理的下一代RAG技术。
anthropics/skills
作为Anthropic官方发布的Agent技能库，该项目持续保持极高热度。这不仅体现了Claude模型在开发者社区中的强大影响力，也揭示了“Agent Skill”作为模块化、可复用的能力单元，已成为扩展Agent功能、构建复杂应用生态的核心模式。
pydantic/monty
该项目使用Rust为AI专门重写了一个安全、轻量的Python解释器，直接应对AI Agent在执行代码时面临的核心安全挑战。这代表了一个重要趋势：随着AI Agent被赋予更大的自主权，其运行环境的安全性、隔离性和资源可控性正成为基础设施层面的关键考量。
Methodology
Data Sources:
- GitHub Trending pages
- GitHub API stargazer timestamps
Metrics:
- Heat_7: Stars gained in last 7 days
- Heat_30: Stars gained in last 30 days
- Score: Weighted scoring combining short-term heat and sustained growth
- Formula: 0.6 × stars_1d + 0.3 × (stars_7d / 7) + 0.1 × (stars_30d / 30)
- Emphasizes recent activity (60%) while considering sustained trends (40%)
Filtering:
- Include keywords: %!s(int=30)
- Exclude keywords: %!s(int=30)
- Categories: %!s(int=30)
Ranking:
- Sort by Heat_30 (descending)
- Tie-break by Score (descending)
- Select top %!d(string=ai, llm, agent, gpt, rag, langchain, machine-learning, deep-learning, neural, transformer, embedding, vector)
Limitations:
- Trending data limited to GitHub's trending algorithm
- Star history may be incomplete for repos with >40k stars (API pagination limits)
- LLM-generated commentary is interpretive, not prescriptive
- Weekly snapshots may miss short-lived trends%!(EXTRA string=tutorial, awesome, learning, course, book, string=infra, agent, llm, rag, vision, int=50)
Generated at: 2026-03-02T00:27:27Z