Report Date: 2026-02-16
Analysis Window: 30 days
Languages Tracked: python, typescript, rust, go, javascript
Top N: 50
Overview
根据2026年2月16日的数据，过去30天内GitHub上AI领域的开源趋势显示，AI Agent项目无论在数量还是社区活跃度上都占据了绝对主导地位。技术栈方面，Python和TypeScript是开发者的首选语言，反映了AI后端能力与前端交互集成的双重热点。
Top 50 Repositories
| Rank | Repository | Category | Language | Heat_7 | Heat_30 | Score |
|---|---|---|---|---|---|---|
| 1 | openclaw | typescript | 0 | 191,849 | 2,067 | |
| 2 | opencode | agent | typescript | 0 | 35,045 | 116 |
| 3 | skills | agent | python | 0 | 28,927 | 96 |
| 4 | shannon | agent | typescript | 10,987 | 18,511 | 532 |
| 5 | agent-skills | agent | javascript | 0 | 15,160 | 50 |
| 6 | claude-mem | agent | typescript | 3,326 | 14,703 | 191 |
| 7 | oh-my-opencode | agent | typescript | 0 | 14,698 | 48 |
| 8 | langextract | llm | python | 7,964 | 11,965 | 381 |
| 9 | pi-mono | agent | typescript | 4,244 | 10,495 | 216 |
| 10 | PageIndex | rag | python | 0 | 9,831 | 32 |
| 11 | dexter | agent | typescript | 3,258 | 8,341 | 167 |
| 12 | beads | agent | go | 0 | 6,093 | 20 |
| 13 | memU | agent | python | 0 | 4,668 | 15 |
| 14 | goose | agent | rust | 0 | 4,628 | 15 |
| 15 | codex | agent | rust | 987 | 4,580 | 57 |
| 16 | agent-lightning | agent | python | 0 | 4,536 | 15 |
| 17 | UltraRAG | rag | python | 0 | 2,804 | 9 |
| 18 | BitNet | llm | python | 0 | 2,795 | 9 |
| 19 | ChatLab | agent | typescript | 0 | 2,661 | 8 |
| 20 | kimi-cli | agent | python | 0 | 2,605 | 8 |
| 21 | claudian | llm | typescript | 0 | 1,575 | 28 |
| 22 | ralph-orchestrator | agent | rust | 0 | 1,156 | 3 |
| 23 | ext-apps | typescript | 0 | 1,140 | 3 | |
| 24 | claude-skills | llm | python | 0 | 1,135 | 3 |
| 25 | worktrunk | agent | rust | 0 | 961 | 3 |
| 26 | screenpipe | rust | 0 | 523 | 1 | |
| 27 | transformer-explainer | llm | javascript | 0 | 341 | 1 |
| 28 | agent-sandbox | agent | go | 0 | 285 | 0 |
| 29 | rowboat | typescript | 0 | 0 | 481 | |
| 30 | chrome-devtools-mcp | agent | typescript | 1,767 | 0 | 289 |
| 31 | free-llm-api-resources | infra | python | 2,925 | 0 | 254 |
| 32 | gh-aw | agent | go | 2,106 | 0 | 218 |
| 33 | monty | rust | 3,113 | 0 | 133 | |
| 34 | unstract | llm | python | 0 | 0 | 129 |
| 35 | letta-code | agent | typescript | 0 | 0 | 115 |
| 36 | RAG-Anything | rag | python | 627 | 0 | 101 |
| 37 | Personal_AI_Infrastructure | agent | typescript | 2,102 | 0 | 90 |
| 38 | open-webui | python | 0 | 0 | 70 | |
| 39 | nanoGPT | llm | python | 538 | 0 | 63 |
| 40 | PocketFlow | agent | python | 0 | 0 | 44 |
| 41 | crush | agent | go | 0 | 0 | 37 |
| 42 | crawl4ai | llm | python | 604 | 0 | 25 |
| 43 | slime | llm | python | 468 | 0 | 20 |
| 44 | chroma | rag | rust | 0 | 0 | 12 |
| 45 | mistral-vibe | agent | python | 0 | 0 | 10 |
| 46 | postiz-app | typescript | 0 | 0 | 10 | |
| 47 | ruvector | rag | rust | 0 | 0 | 9 |
| 48 | baserow | agent | python | 193 | 0 | 8 |
| 49 | claude-skills | llm | typescript | 185 | 0 | 7 |
| 50 | XcodeBuildMCP | agent | typescript | 147 | 0 | 6 |
Category Breakdown
llm (10 projects)
LLM（大语言模型）相关项目是第二大类别，拥有10个项目，并保持着较高的平均热度，表明围绕基础模型的工具和应用开发依然强劲。
Average Heat_7: 975
Average Score: 66
- langextract - A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
- BitNet - Official inference framework for 1-bit LLMs
- claudian - An Obsidian plugin that embeds Claude Code as an AI collaborator in your vault
- claude-skills - A Collection of Skills for Claude Code and Claude AI for real-world Usage. Including Claude Code Subagents, Claude Code Commnads
- transformer-explainer - Transformer Explained Visually: Learn How LLM Transformer Models Work with Interactive Visualization
- unstract - No-code LLM Platform to launch APIs and ETL Pipelines to structure unstructured documents
- nanoGPT - The simplest, fastest repository for training/finetuning medium-sized GPTs.
- crawl4ai - 🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN
- slime - slime is an LLM post-training framework for RL Scaling.
- claude-skills - Skills for Claude Code CLI such as full stack dev Cloudflare, React, Tailwind v4, and AI integrations.
rag (5 projects)
RAG（检索增强生成）领域项目数量不多（5个），平均热度和得分相对较低，说明该方向虽然重要，但近期现象级的新项目较少。
Average Heat_7: 125
Average Score: 32
- PageIndex - 📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG
- UltraRAG - A Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines
- RAG-Anything - "RAG-Anything: All-in-One RAG Framework"
- chroma - Open-source search and retrieval database for AI applications.
- ruvector - A distributed vector database that learns. Store embeddings, query with Cypher, scale horizontally with Raft consensus, and let the index improve itself through Graph Neural Networks.
infra (1 projects)
基础设施（infra）类别仅有1个项目，但其热度和得分数据极高，这由单个高价值资源集合类项目（free-llm-api-resources）拉动，属于异常值。
Average Heat_7: 2,925
Average Score: 254
- free-llm-api-resources - A list of free LLM inference resources accessible via API.
(7 projects)
未分类项目（7个）的平均得分（395）显著高于其他类别，这可能意味着一些具有高影响力的新兴项目或跨领域工具（如排名第一的openclaw）难以被简单归类。
Average Heat_7: 444
Average Score: 395
- openclaw - Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
- ext-apps - Official repo for spec & SDK of MCP Apps protocol - standard for UIs embedded AI chatbots, served by MCP servers
- screenpipe - screenpipe turns your computer into a personal AI that knows everything you've done. record. search. automate. all local, all private, all yours.
- rowboat - Open-source AI coworker, with memory
- monty - A minimal, secure Python interpreter written in Rust for use by AI
- open-webui - User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
- postiz-app - 📨 The ultimate social media scheduling tool, with a bunch of AI 🤖
agent (27 projects)
AI Agent是本周期的绝对焦点，项目数量（27个）和平均热度（1078.4）均遥遥领先，显示出开发者社区在该领域的极高投入和活跃度。
Average Heat_7: 1,078
Average Score: 88
- opencode - The open source coding agent.
- skills - Public repository for Agent Skills
- 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.
- agent-skills - Vercel's official collection of agent skills
- 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 ...
- oh-my-opencode - the best agent harness
- 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
- beads - Beads - A memory upgrade for your coding agent
- 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。
- 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
- worktrunk - Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows
- agent-sandbox - agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes.
- chrome-devtools-mcp - Chrome DevTools for coding agents
- gh-aw - GitHub Agentic Workflows
- letta-code - The memory-first coding agent
- Personal_AI_Infrastructure - Agentic AI Infrastructure for magnifying HUMAN capabilities.
- PocketFlow - Pocket Flow: 100-line LLM framework. Let Agents build Agents!
- crush - Glamourous agentic coding for all 💘
- mistral-vibe - Minimal CLI coding agent by Mistral
- baserow - Build databases, automations, apps & agents with AI — no code. Open source platform available on cloud and self-hosted. GDPR, HIPAA, SOC 2 compliant. Best Airtable alternative.
- XcodeBuildMCP - A Model Context Protocol (MCP) server and CLI that provides tools for agent use when working on iOS and macOS projects.
Dark Horse Projects
黑马项目中出现了得分（score）极高的项目，如 openclaw (2067分) 和 rowboat (481分)，即使它们近7日热度较低，也表明其已累积了大量的社区关注。另一些项目如 KeygraphHQ/shannon 和 google/langextract 则兼 具高分和近期高热度，显示出持续增长的强劲潜力。
| Repository | Score | Heat_30 | Heat_7 | Category |
|---|---|---|---|---|
| openclaw | 2,067 | 191,849 | 0 | |
| opencode | 116 | 35,045 | 0 | agent |
| shannon | 532 | 18,511 | 10,987 | agent |
| claude-mem | 191 | 14,703 | 3,326 | agent |
| langextract | 381 | 11,965 | 7,964 | llm |
| pi-mono | 216 | 10,495 | 4,244 | agent |
| dexter | 167 | 8,341 | 3,258 | agent |
| rowboat | 481 | 0 | 0 | |
| chrome-devtools-mcp | 289 | 0 | 1,767 | agent |
| free-llm-api-resources | 254 | 0 | 2,925 | infra |
| gh-aw | 218 | 0 | 2,106 | agent |
| monty | 133 | 0 | 3,113 | |
| unstract | 129 | 0 | 0 | llm |
| letta-code | 115 | 0 | 0 | agent |
| RAG-Anything | 101 | 0 | 627 | rag |
Consecutive Appearances
上榜项目中存在大量的“常客”（29个），许多核心项目（如 openclaw, opencode, shannon）已连续4周上榜。这表明AI Agent领域的创新项目能够持续吸引社区关注，而非昙花一现。同时，部分项目近期热度虽有回落，但仍凭借前期积累的优势保持在榜单前列。
| Repository | Weeks in Top | Category | Current Heat_7 |
|---|---|---|---|
| openclaw | 4 | 0 | |
| opencode | 4 | agent | 0 |
| skills | 4 | agent | 0 |
| shannon | 4 | agent | 10,987 |
| claude-mem | 4 | agent | 3,326 |
| oh-my-opencode | 4 | agent | 0 |
| langextract | 2 | llm | 7,964 |
| pi-mono | 4 | agent | 4,244 |
| PageIndex | 4 | rag | 0 |
| dexter | 4 | agent | 3,258 |
| beads | 4 | agent | 0 |
| memU | 4 | agent | 0 |
| goose | 4 | agent | 0 |
| codex | 4 | agent | 987 |
| agent-lightning | 4 | agent | 0 |
| UltraRAG | 4 | rag | 0 |
| BitNet | 4 | llm | 0 |
| ChatLab | 4 | agent | 0 |
| kimi-cli | 4 | agent | 0 |
| ralph-orchestrator | 4 | agent | 0 |
| ext-apps | 4 | 0 | |
| claude-skills | 4 | llm | 0 |
| worktrunk | 4 | agent | 0 |
| transformer-explainer | 2 | llm | 0 |
| gh-aw | 2 | agent | 2,106 |
| monty | 2 | 3,113 | |
| nanoGPT | 2 | llm | 538 |
| crush | 4 | agent | 0 |
| chroma | 2 | rag | 0 |
Highlighted Repositories
KeygraphHQ/shannon
该项目是一个专注于网络安全领域的自主AI黑客，其明确的性能指标（在XBOW基准测试中成功率96.15%）和持续的高热度（近7日热度10987）表明，具备可验证能力的垂直领域AI Agent是当前的一大热点。连续4周上榜也证明了其强大的生命力。
google/langextract
由谷歌推出的这个Python库，专注于从非结构化文本中提取结构化信息，并提供精确的来源追溯和可视化功能。这直接解决了LLM应用中的一个核心痛点——结果的可解释性与可靠性，因此获得了很高的社区关注。
openclaw/openclaw
作为榜首项目，openclaw 拥有极高的综合得分（2067）和30日热度，定位为个人AI助理。尽管其近7日热度为零，暗示其热度峰值可能已经过去，但其庞大的累计关注度说明“个人AI助理”这一概念具有巨大的市场吸引力。
pydantic/monty
该项目由知名开源组织Pydantic推出，是一个用Rust编写的、专为AI设计的安全Python解释 器。它反映了一个重要的底层技术趋势：随着AI Agent执行代码能力的增强，社区对安全性和沙箱环境的需求日益增长。
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=agent, llm, rag, vision, infra, int=50)
Generated at: 2026-02-16T00:28:44Z