Report Date: 2026-04-27
Analysis Window: 30 days
Languages Tracked: python, typescript, rust, go, javascript
Top N: 50
Overview
根据最近30天的数据，GitHub上的AI领域展现出极高的活跃度，其中 'Agent' (智能体) 已成为最热门和项目最集中的赛道。开发者们高度关注AI编码助手和 相关生态的构建，同时，尽管Python和TypeScript仍是主流，但Rust语言因其高性能特质在AI工具链中正迅速崛起。
Top 50 Repositories
| Rank | Repository | Category | Language | Heat_7 | Heat_30 | Score |
|---|---|---|---|---|---|---|
| 1 | hermes-agent | agent | python | 0 | 104,773 | 349 |
| 2 | everything-claude-code | agent | javascript | 0 | 59,272 | 197 |
| 3 | claude-howto | agent | python | 0 | 27,992 | 93 |
| 4 | claude-mem | agent | typescript | 4,967 | 27,436 | 304 |
| 5 | oh-my-codex | agent | typescript | 0 | 23,592 | 78 |
| 6 | rtk | llm | rust | 6,250 | 22,248 | 342 |
| 7 | VibeVoice | python | 0 | 17,850 | 59 | |
| 8 | codex | agent | rust | 0 | 10,717 | 151 |
| 9 | onyx | llm | python | 0 | 10,544 | 35 |
| 10 | goose | agent | rust | 0 | 9,965 | 33 |
| 11 | impeccable | javascript | 0 | 8,438 | 28 | |
| 12 | marketingskills | llm | javascript | 2,821 | 8,393 | 148 |
| 13 | ai-hedge-fund | python | 0 | 8,077 | 26 | |
| 14 | GenericAgent | agent | python | 2,936 | 6,534 | 147 |
| 15 | Archon | typescript | 0 | 6,105 | 20 | |
| 16 | new-api | llm | go | 0 | 6,058 | 20 |
| 17 | Open-Generative-AI | vision | javascript | 3,448 | 6,030 | 516 |
| 18 | claude-skills | llm | python | 0 | 5,795 | 19 |
| 19 | ppt-master | vision | python | 1,850 | 5,275 | 96 |
| 20 | evolver | agent | javascript | 1,512 | 5,248 | 82 |
| 21 | fff.nvim | agent | rust | 0 | 4,316 | 14 |
| 22 | RAG-Anything | rag | python | 2,639 | 4,201 | 127 |
| 23 | claudian | llm | typescript | 0 | 4,101 | 13 |
| 24 | Pixelle-Video | python | 0 | 3,630 | 12 | |
| 25 | AI-Scientist-v2 | agent | python | 0 | 3,412 | 11 |
| 26 | qwen-code | agent | typescript | 0 | 3,052 | 10 |
| 27 | anything-llm | javascript | 0 | 2,465 | 8 | |
| 28 | camofox-browser | agent | javascript | 0 | 2,293 | 100 |
| 29 | OmniRoute | llm | typescript | 0 | 2,278 | 7 |
| 30 | 9router | llm | javascript | 489 | 1,916 | 27 |
| 31 | forgecode | llm | rust | 0 | 1,868 | 6 |
| 32 | honcho | agent | python | 0 | 1,838 | 6 |
| 33 | one-api | llm | javascript | 0 | 1,699 | 5 |
| 34 | memvid | rag | rust | 0 | 1,550 | 5 |
| 35 | bifrost | go | 381 | 1,095 | 19 | |
| 36 | clawpanel | agent | javascript | 0 | 968 | 3 |
| 37 | free-coding-models | llm | javascript | 0 | 826 | 2 |
| 38 | microsandbox | agent | rust | 280 | 705 | 14 |
| 39 | tabby | rust | 0 | 522 | 1 | |
| 40 | pi-skills | agent | javascript | 120 | 466 | 6 |
| 41 | mistral.rs | infra | rust | 0 | 345 | 5 |
| 42 | pi-mono | agent | typescript | 3,173 | 0 | 450 |
| 43 | GitNexus | agent | typescript | 0 | 0 | 420 |
| 44 | openclaw | typescript | 0 | 0 | 376 | |
| 45 | opencode | agent | typescript | 0 | 0 | 307 |
| 46 | multica | agent | typescript | 4,882 | 0 | 209 |
| 47 | posthog | python | 0 | 0 | 202 | |
| 48 | ruflo | agent | typescript | 0 | 0 | 153 |
| 49 | claude-context | agent | typescript | 3,537 | 0 | 151 |
| 50 | TrendRadar | python | 3,153 | 0 | 135 |
Category Breakdown
agent (23 projects)
智能体是绝对的主导类别，项目数量（23个）遥遥领先，表明该方向是当前AI开源创新的核心。
Average Heat_7: 930
Average Score: 142
- hermes-agent - The agent that grows with you
- everything-claude-code - The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
- claude-howto - A visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
- 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-codex - OmX - Oh My codeX: Your codex is not alone. Add hooks, agent teams, HUDs, and so much more.
- codex - Lightweight coding agent that runs in your terminal
- goose - an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
- GenericAgent - Self-evolving agent: grows skill tree from 3.3K-line seed, achieving full system control with 6x less token consumption
- evolver - The GEP-powered self-evolving engine for AI agents. Auditable evolution with Genes, Capsules, and Events. | evomap.ai
- fff.nvim - The fastest and the most accurate file search toolkit for AI agents, Neovim, Rust, C, and NodeJS
- AI-Scientist-v2 - The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
- qwen-code - An open-source AI agent that lives in your terminal.
- camofox-browser - Stealth headless browser for AI agents — bypass Cloudflare, bot detection, and anti-scraping. Drop-in Puppeteer/Playwright replacement.
- honcho - Memory library for building stateful agents
- clawpanel - 🦞 OpenClaw & Hermes Agent 多引擎 AI 管理面板 — 内置 AI 助手（工具调用 + 图片识别 + 多模态），一键安装 | Tauri v2 跨平台桌面应用 | 11 种语言
- microsandbox - 🧱 secure, local and programmable sandboxes for AI agents
- pi-skills - Skills for pi coding agent (compatible with Claude Code and Codex CLI)
- pi-mono - AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods
- 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 ...
- opencode - The open source coding agent.
- multica - The open-source managed agents platform. Turn coding agents into real teammates — assign tasks, track progress, compound 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...
- claude-context - Code search MCP for Claude Code. Make entire codebase the context for any coding agent.
llm (11 projects)
大语言模型作为基础技术，项目数量位居第二，但平均热度相对Agent较低，显示出市场焦点正从模型本身转向应用层。
Average Heat_7: 869
Average Score: 56
- rtk - CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
- onyx - Open Source AI Platform - AI Chat with advanced features that works with every LLM
- marketingskills - Marketing skills for Claude Code and AI agents. CRO, copywriting, SEO, analytics, and growth engineering.
- new-api - A unified AI model hub for aggregation & distribution. It supports cross-converting various LLMs into OpenAI-compatible, Claude-compatible, or Gemini-compatible formats. A centralized gateway for p...
- claude-skills - 232+ Claude Code skills & agent plugins for Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents — engineering, marketing, product, compliance, C-level advisory.
- claudian - An Obsidian plugin that embeds Claude Code as an AI collaborator in your vault
- OmniRoute - OmniRoute is an AI gateway for multi-provider LLMs: an OpenAI-compatible endpoint with smart routing, load balancing, retries, and fallbacks. Add policies, rate limits, caching, and observability f...
- 9router - Connect All AI Code Tools (Claude Code, Cursor, Antigravity, Copilot, Codex, Gemini, OpenCode, Cline, OpenClaw...) to 40+ AI Providers & 100+ Models
- forgecode - AI enabled pair programmer for Claude, GPT, O Series, Grok, Deepseek, Gemini and 300+ models
- one-api - LLM API 管理 & 分发系统，支持 OpenAI、Azure、Anthropic Claude、Google Gemini、DeepSeek、字节豆包、ChatGLM、文心一言、讯飞星火、通义千问、360 智脑、腾讯混元...
- free-coding-models - Find, benchmark and install in CLI 200+ FREE coding LLM models across 20+ providers in real time
(11 projects)
Average Heat_7: 321
Average Score: 80
- VibeVoice - Open-Source Frontier Voice AI
- impeccable - The design language that makes your AI harness better at design.
- ai-hedge-fund - An AI Hedge Fund Team
- Archon - The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.
- Pixelle-Video - 🚀 AI 全自动短视频引擎 | AI Fully Automated Short Video Engine
- anything-llm - The all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
- bifrost - Fastest enterprise AI gateway (50x faster than LiteLLM) with adaptive load balancer, cluster mode, guardrails, 1000+ models support & <100 µs overhead at 5k RPS.
- tabby - Self-hosted AI coding assistant
- openclaw - Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
- posthog - 🦔 PostHog is an all-in-one developer platform for building successful products. We offer product analytics, web analytics, session replay, error tracking, feature flags, experimentation, surveys...
- TrendRadar - ⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts.🎯 告别信息过载，你的 AI 舆情监控助手与热点筛选工具！聚合多平台热�...
vision (2 projects)
视觉类项目虽少（仅2个），却拥有极高的平均热度和评分，说明该领域一旦出现突破性项目，便能迅速吸引大量关注。
Average Heat_7: 2,649
Average Score: 306
- Open-Generative-AI - Uncensored, open-source alternative to Higgsfield AI, Freepik AI, Krea AI, Openart AI — Free, unrestricted AI image & video generation studio with 200+ models (Flux, Midjourney, Kling, Sora, Veo)...
- ppt-master - AI generates natively editable PPTX from any document — real PowerPoint shapes, not images — no design skills needed
rag (2 projects)
与Vision类似，RAG（检索增强生成）项目数量不多，但热度很高，反映了市场对高效、可靠的RAG框架的持续渴求。
Average Heat_7: 1,319
Average Score: 66
- RAG-Anything - "RAG-Anything: All-in-One RAG Framework"
- 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.
infra (1 projects)
基础设施类别项目稀少且热度较低，可能意味着该领域的创新更集中在少数成熟项目上，新项目难以突围。
Average Heat_7: 0
Average Score: 5
- mistral.rs - Fast, flexible LLM inference
Dark Horse Projects
本期黑马项目（高分项目）呈现出两种主要模式：一种是像hermes-agent那样在30天内积累了巨大热度但近期热度回落的成熟项目；另一种则是如pi-mono和multica这样近期（7天内）才爆发式增长的新星。这表明，一个项目的影响力（score）并不仅仅由短期热度决定，其长期积累或爆发潜力同样重要。
| Repository | Score | Heat_30 | Heat_7 | Category |
|---|---|---|---|---|
| hermes-agent | 349 | 104,773 | 0 | agent |
| everything-claude-code | 197 | 59,272 | 0 | agent |
| claude-mem | 304 | 27,436 | 4,967 | agent |
| rtk | 342 | 22,248 | 6,250 | llm |
| codex | 151 | 10,717 | 0 | agent |
| marketingskills | 148 | 8,393 | 2,821 | llm |
| GenericAgent | 147 | 6,534 | 2,936 | agent |
| Open-Generative-AI | 516 | 6,030 | 3,448 | vision |
| RAG-Anything | 127 | 4,201 | 2,639 | rag |
| camofox-browser | 100 | 2,293 | 0 | agent |
| pi-mono | 450 | 0 | 3,173 | agent |
| GitNexus | 420 | 0 | 0 | agent |
| openclaw | 376 | 0 | 0 | |
| opencode | 307 | 0 | 0 | agent |
| multica | 209 | 0 | 4,882 | agent |
| posthog | 202 | 0 | 0 | |
| ruflo | 153 | 0 | 0 | agent |
| claude-context | 151 | 0 | 3,537 | agent |
| TrendRadar | 135 | 0 | 3,153 |
Consecutive Appearances
榜单上出现了大量“常客”项目，如已上榜14周的openai/codex，这说明AI领域虽然日新月异，但一些核心的、高质量的项目能够持续保持其影响力和社区关注度。这些项目的存在为开发者提供了一个相对稳定的技术基石。
| Repository | Weeks in Top | Category | Current Heat_7 |
|---|---|---|---|
| hermes-agent | 8 | agent | 0 |
| everything-claude-code | 6 | agent | 0 |
| claude-mem | 3 | agent | 4,967 |
| VibeVoice | 5 | 0 | |
| codex | 14 | agent | 0 |
| onyx | 5 | llm | 0 |
| goose | 3 | agent | 0 |
| impeccable | 8 | 0 | |
| marketingskills | 7 | llm | 2,821 |
| ai-hedge-fund | 8 | 0 | |
| GenericAgent | 2 | agent | 2,936 |
| Archon | 3 | 0 | |
| new-api | 5 | llm | 0 |
| Open-Generative-AI | 3 | vision | 3,448 |
| claude-skills | 6 | llm | 0 |
| evolver | 2 | agent | 1,512 |
| fff.nvim | 5 | agent | 0 |
| claudian | 3 | llm | 0 |
| AI-Scientist-v2 | 2 | agent | 0 |
| qwen-code | 2 | agent | 0 |
| anything-llm | 2 | 0 | |
| camofox-browser | 3 | agent | 0 |
| OmniRoute | 2 | llm | 0 |
| 9router | 6 | llm | 489 |
| forgecode | 3 | llm | 0 |
| one-api | 7 | llm | 0 |
| memvid | 4 | rag | 0 |
| bifrost | 4 | 381 | |
| pi-mono | 2 | agent | 3,173 |
| multica | 3 | agent | 4,882 |
| ruflo | 2 | agent | 0 |
Highlighted Repositories
NousResearch/hermes-agent
作为榜首项目，'hermes-agent' 是智能体领域的标杆。数据显示其在30天内积累了极高的热度（104773），但近7天热度为0，这可能暗示其近期没有重大更新，但其长期积累的影响力依然使其稳坐榜首。
rtk-ai/rtk
该项目是Rust在AI领域兴起的典型代表。它作为一个命令行工具，精准解决了开发者在使用LLM时遇到的“token消耗”痛点，其实用性和高性能特性使其获得了很高的评分（342）。
Anil-matcha/Open-Generative-AI
这个视觉类项目以其极高的评分（516）成为本期最亮眼的黑马之一。它明确地将自己定位为多个商业AI平台的开源、无审查替代品，并支持自托管，这精准地抓住了开源社区对数据隐私和创作自由的核心诉求。
badlogic/pi-mono
此项目虽然总排名不高（42），但其极高的评分（450）和近7天内爆发式的热度（3173）显示出巨大的潜力。它提供了一个全面的AI智能体工具包，而非单一功能，预示着AI开发正从单一工具向集成化平台演进。
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: 2026-04-27T00:39:08Z