Report Date: 2026-02-09
Analysis Window: 30 days
Languages Tracked: python, typescript, rust, go, javascript
Top N: 50
Overview
根据近30天的数据，GitHub AI领域的开源趋势显示，AI Agent项目占据绝对主导地位，在前50个项目中占据了27席。该领域创新活跃，有49个项目是新近创建的，显示出极快的迭代速度。在技术栈方面，TypeScript和Python是开发者最青睐的编程语言。
Top 50 Repositories
| Rank | Repository | Category | Language | Heat_7 | Heat_30 | Score |
|---|---|---|---|---|---|---|
| 1 | openclaw | typescript | 38,587 | 173,310 | 2,231 | |
| 2 | opencode | agent | typescript | 0 | 45,981 | 153 |
| 3 | skills | agent | python | 0 | 30,302 | 101 |
| 4 | oh-my-opencode | agent | typescript | 0 | 17,955 | 59 |
| 5 | claude-mem | agent | typescript | 9,415 | 13,110 | 447 |
| 6 | PageIndex | rag | python | 2,712 | 9,236 | 147 |
| 7 | shannon | agent | typescript | 7,330 | 8,495 | 2,429 |
| 8 | vibe-kanban | llm | typescript | 0 | 6,888 | 22 |
| 9 | UI-TARS-desktop | agent | typescript | 2,337 | 6,847 | 122 |
| 10 | pi-mono | agent | typescript | 3,570 | 6,737 | 175 |
| 11 | beads | agent | go | 0 | 6,606 | 22 |
| 12 | dexter | agent | typescript | 2,796 | 6,309 | 764 |
| 13 | memU | agent | python | 1,612 | 4,846 | 85 |
| 14 | goose | agent | rust | 0 | 4,581 | 15 |
| 15 | codex | agent | rust | 1,310 | 4,396 | 70 |
| 16 | agent-lightning | agent | python | 1,456 | 4,171 | 76 |
| 17 | context7 | llm | typescript | 0 | 3,981 | 13 |
| 18 | ChatLab | agent | typescript | 0 | 2,993 | 9 |
| 19 | UltraRAG | rag | python | 0 | 2,761 | 9 |
| 20 | crush | agent | go | 0 | 2,682 | 8 |
| 21 | BitNet | llm | python | 0 | 2,626 | 8 |
| 22 | kimi-cli | agent | python | 0 | 2,446 | 8 |
| 23 | supermemory | typescript | 0 | 2,298 | 7 | |
| 24 | memvid | rag | rust | 0 | 1,426 | 4 |
| 25 | ralph-orchestrator | agent | rust | 0 | 1,337 | 4 |
| 26 | OpenAgentsControl | agent | typescript | 0 | 1,192 | 3 |
| 27 | claude-skills | llm | python | 0 | 1,149 | 3 |
| 28 | ext-apps | typescript | 0 | 1,088 | 3 | |
| 29 | worktrunk | agent | rust | 0 | 962 | 3 |
| 30 | chroma | rag | rust | 0 | 714 | 2 |
| 31 | meeting-minutes | rust | 0 | 655 | 2 | |
| 32 | bifrost | llm | go | 0 | 527 | 1 |
| 33 | monty | rust | 0 | 0 | 273 | |
| 34 | langextract | llm | python | 0 | 0 | 262 |
| 35 | transformer-explainer | llm | javascript | 160 | 0 | 58 |
| 36 | agents | agent | python | 0 | 0 | 57 |
| 37 | Maestro | agent | typescript | 1,181 | 0 | 50 |
| 38 | airweave | agent | python | 0 | 0 | 37 |
| 39 | easy-dataset | llm | javascript | 222 | 0 | 33 |
| 40 | nanoGPT | llm | python | 0 | 0 | 24 |
| 41 | cognee | agent | python | 567 | 0 | 24 |
| 42 | SparkyFitness | typescript | 0 | 0 | 23 | |
| 43 | plano | agent | rust | 0 | 0 | 18 |
| 44 | RD-Agent | python | 359 | 0 | 15 | |
| 45 | LocalAI | vision | go | 0 | 0 | 15 |
| 46 | gh-aw | agent | go | 85 | 0 | 12 |
| 47 | UI-TARS | agent | python | 239 | 0 | 10 |
| 48 | axonhub | llm | go | 210 | 0 | 9 |
| 49 | Acontext | agent | typescript | 0 | 0 | 4 |
Category Breakdown
llm (10 projects)
大语言模型（LLM）领域依然是重要阵地（10个项目），但热度相对Agent较低，新项目多集中在模型应用、数据处理和推理优化等工具层面。
Average Heat_7: 59
Average Score: 43
- vibe-kanban - Get 10X more out of Claude Code, Codex or any coding agent
- context7 - Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
- BitNet - Official inference framework for 1-bit LLMs
- claude-skills - A Collection of Skills for Claude Code and Claude AI for real-world Usage. Including Claude Code Subagents, Claude Code Commnads
- bifrost - Fastest LLM gateway (50x faster than LiteLLM) with adaptive load balancer, cluster mode, guardrails, 1000+ models support & <100 µs overhead at 5k RPS.
- langextract - A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
- transformer-explainer - Transformer Explained Visually: Learn How LLM Transformer Models Work with Interactive Visualization
- easy-dataset - A powerful tool for creating datasets for LLM fine-tuning 、RAG and Eval
- nanoGPT - The simplest, fastest repository for training/finetuning medium-sized GPTs.
- axonhub - ⚡️ Open-source AI Gateway — Use any SDK to call 100+ LLMs. Built-in failover, load balancing, cost control & end-to-end tracing.
vision (1 projects)
计算机视觉（Vision）领域在本次趋势榜中表现较为边缘化，仅有1个项目上榜，关注度较低。
Average Heat_7: 0
Average Score: 15
- LocalAI - 🤖 The free, Open Source alternative to OpenAI, Claude and others. Self-hosted and local-first. Drop-in replacement, running on consumer-grade hardware. No GPU required. Runs gguf, transformers, ...
(7 projects)
少数未分类项目（7个）贡献了极高的平均热度，这主要是由排名第一的 openclaw/openclaw 等个别爆款项目拉动所致。
Average Heat_7: 5,563
Average Score: 364
- openclaw - Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
- supermemory - Memory engine and app that is extremely fast, scalable. The Memory API for the AI era.
- ext-apps - Official repo for spec & SDK of MCP Apps protocol - standard for UIs embedded AI chatbots, served by MCP servers
- meeting-minutes - 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...
- monty - A minimal, secure Python interpreter written in Rust for use by AI
- SparkyFitness - SparkyFitness: Built for Families. Powered by AI. Track food, fitness, water, and health — together.
- RD-Agent - Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are commi...
agent (27 projects)
Agent是当前最热门的研发方向，项目数量（27个）和平均热度均遥遥领先，覆盖了从编码、研究到安全测试等多个应用场景。
Average Heat_7: 1,181
Average Score: 176
- opencode - The open source coding agent.
- skills - Public repository for Agent Skills
- oh-my-opencode - the best agent harness
- 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 ...
- 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.
- UI-TARS-desktop - The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra
- pi-mono - AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods
- beads - Beads - A memory upgrade for your coding agent
- dexter - An autonomous agent for deep financial research
- memU - Memory for 24/7 proactive agents like openclaw (moltbot, clawdbot).
- goose - an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
- codex - Lightweight coding agent that runs in your terminal
- agent-lightning - The absolute trainer to light up AI agents.
- ChatLab - A Local-first chat analysis tool: Relive your social memories powered by SQL and AI Agents. 本地化的聊天记录分析工具，通过 SQL 和 AI Agent 回顾你的社交记忆。
- crush - Glamourous agentic coding for all 💘
- kimi-cli - Kimi Code CLI is your next CLI agent.
- ralph-orchestrator - An improved implementation of the Ralph Wiggum technique for autonomous AI agent orchestration
- OpenAgentsControl - AI agent framework for plan-first development workflows with approval-based execution. Multi-language support (TypeScript, Python, Go, Rust) with automatic testing, code review, and validation buil...
- worktrunk - Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows
- agents - Intelligent automation and multi-agent orchestration for Claude Code
- Maestro - Agent Orchestration Command Center
- airweave - Open-source context retrieval layer for AI agents
- cognee - Memory for AI Agents in 6 lines of code
- plano - Delivery infrastructure for agentic apps - Plano is an AI-native proxy and data plane that offloads plumbing work, so you stay focused on your agent's core logic (via any AI framework).
- gh-aw - GitHub Agentic Workflows
- UI-TARS - Pioneering Automated GUI Interaction with Native Agents
- Acontext - Context Data Platform for AI Agents
rag (4 projects)
检索增强生成（RAG）作为一个关键技术方向（4个项目），保持着一定的关注度，但整体项目数量和平均得分相对不高，显示出该领域仍在探索更成熟的解决方案。
Average Heat_7: 678
Average Score: 40
- PageIndex - 📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG
- UltraRAG - UltraRAG v3: A Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines
- memvid - Memory layer for AI Agents. Replace complex RAG pipelines with a serverless, single-file memory layer. Give your agents instant retrieval and long-term memory.
- chroma - Open-source search and retrieval database for AI applications.
Dark Horse Projects
本期榜单中出现了多匹“黑马”项目，它们凭借高分（score）获得了社区的高度认可。其中，openclaw/openclaw（得分2231）和KeygraphHQ/shannon（得分2429）不仅得分高，近期热度（heat_7）也非常活跃。另一些项目如pydantic/monty（得分273）和google/langextract（得分262）近期热度为零但得分很高，这可能归功于其背后知名组织的影响力和项目本身的基础价值。
| Repository | Score | Heat_30 | Heat_7 | Category |
|---|---|---|---|---|
| openclaw | 2,231 | 173,310 | 38,587 | |
| opencode | 153 | 45,981 | 0 | agent |
| skills | 101 | 30,302 | 0 | agent |
| claude-mem | 447 | 13,110 | 9,415 | agent |
| PageIndex | 147 | 9,236 | 2,712 | rag |
| shannon | 2,429 | 8,495 | 7,330 | agent |
| UI-TARS-desktop | 122 | 6,847 | 2,337 | agent |
| pi-mono | 175 | 6,737 | 3,570 | agent |
| dexter | 764 | 6,309 | 2,796 | agent |
| monty | 273 | 0 | 0 | |
| langextract | 262 | 0 | 0 | llm |
Consecutive Appearances
大量项目（数据中列出38个）连续3周上榜，表明这些项目不仅仅是昙花一现，而是获得了持续的关注。值得注意的是，许多连续上榜的项目近期热度（current_heat_7）已降至零，这说明它们的热度高峰期可能已过，但凭借过去30天的累积影响力依然停留在榜单上。
| Repository | Weeks in Top | Category | Current Heat_7 |
|---|---|---|---|
| openclaw | 3 | 38,587 | |
| opencode | 3 | agent | 0 |
| skills | 3 | agent | 0 |
| oh-my-opencode | 3 | agent | 0 |
| claude-mem | 3 | agent | 9,415 |
| PageIndex | 3 | rag | 2,712 |
| shannon | 3 | agent | 7,330 |
| vibe-kanban | 3 | llm | 0 |
| UI-TARS-desktop | 3 | agent | 2,337 |
| pi-mono | 3 | agent | 3,570 |
| beads | 3 | agent | 0 |
| dexter | 3 | agent | 2,796 |
| memU | 3 | agent | 1,612 |
| goose | 3 | agent | 0 |
| codex | 3 | agent | 1,310 |
| agent-lightning | 3 | agent | 1,456 |
| context7 | 3 | llm | 0 |
| ChatLab | 3 | agent | 0 |
| UltraRAG | 3 | rag | 0 |
| crush | 3 | agent | 0 |
| BitNet | 3 | llm | 0 |
| kimi-cli | 3 | agent | 0 |
| supermemory | 3 | 0 | |
| memvid | 3 | rag | 0 |
| ralph-orchestrator | 3 | agent | 0 |
| OpenAgentsControl | 3 | agent | 0 |
| claude-skills | 3 | llm | 0 |
| ext-apps | 3 | 0 | |
| worktrunk | 3 | agent | 0 |
| meeting-minutes | 3 | 0 | |
| bifrost | 3 | llm | 0 |
| Maestro | 3 | agent | 1,181 |
| easy-dataset | 3 | llm | 222 |
| cognee | 3 | agent | 567 |
| plano | 3 | agent | 0 |
| RD-Agent | 3 | 359 | |
| UI-TARS | 3 | agent | 239 |
| axonhub | 3 | llm | 210 |
Highlighted Repositories
openclaw/openclaw
作为榜单第一且热度与评分遥遥领先的项目，openclaw 反映了市场对个人化、跨平台AI助理的强烈需求。该项目使用TypeScript构建，体现了用Web技术栈构建复杂AI应用的趋势。其未被明确分类的特点可能暗示它是一个综合性平台，而非单一功能工具。
KeygraphHQ/shannon
shannon 以其在自主网络安全领域的独特应用脱颖而出，专注于利用AI发现应用漏洞。其高达2429的评分表明它在安全和AI社区引起了巨大关注，这也是AI Agent从通用编码向专业化、高价值领域拓展的典型案例。
pydantic/monty
pydantic/monty 排名不高且近期无热度，但其重要性不容忽视。该项目由知名团队Pydantic开发，旨在为AI Agent提供一个安全、轻量级的Python解释器，解决了AI安全执行代码的核心痛点。项目采用Rust编写，也反映了AI基础设施领域对性能和安全性的日益重视。
anomalyco/opencode
opencode 作为榜单第二名，曾是AI编码代理领域的引领者。尽管其近期热度为零，但过去30天的高热度和连续上榜记录证明了其巨大的初期影响力。这种热度变化模式可能表明，该项目在经历爆发式增长后，已进入一个相对稳定的发展阶段。
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: 2026-02-09T00:29:12Z