GitHub Discovery ·
Included in The AI Stack
Find the next viral AI repo before everyone posts about it.
GitHub Discovery searches every new public repo and ranks them by stars per day, so you see what's taking off while it's still early enough to cover, build on or sell.
| # | Repo | Stars | ★/day | Lang | Age |
|---|---|---|---|---|---|
| 1 | feder-cr/dots Open-source dots for the web: an AI agent with its own browser, one that does not get blocked. ai-agent · ai-agents · ai-browser · anti-detect-browser · browser-agent · browser-automation | 1,933 | 1,809.2 | Python | 1d |
| 2 | zai-org/ZCode Z.ai's coding agent harness. Powerful, intelligent, extensible. | 7,253 | 688.8 | TypeScript | 11d |
| 3 | nanaism/yomiyasu AI生成された文章の不自然さを取り除き、人間にとって自然な日本語へ書き直すためのAgent Skill agent-skills · ai-writing · claude-code · codex · cursor · japanese | 258 | 258 | Python | 1d |
| 4 | kaankiziltug/logo-design-skill A comprehensive logo-design skill for Claude, Gemini CLI, Codex and other AI agents: principles, process, SVG craft, testing tools and a 1,400+ logo reference library. agent-skills · branding · claude · claude-skills · codex · gemini-cli | 1,110 | 237.1 | HTML | 5d |
| 5 | lemomo-ai/lemo-opuscar 43 film styles, each a reusable style prompt plus a short film made entirely in code by Claude Opus 5.5. Pick a style, bring your own story, and let your agent direct. | Opus5.5 x 43 种影片风格:风格提示词 + 纯代码样片 + 导演与技术指南 ai-agents · ai-video · animation · canvas · claude · claude-code | 683 | 148.3 | JavaScript | 5d |
| 6 | shadcn-ui/lint An agent-first linter for Tailwind design systems. Write design system rules that agents can verify. agents · ai · design · design-system · design-tools · shadcn | 2,976 | 104.3 | TypeScript | 29d |
| 7 | openJiuwen-ai/iCode A lightweight, extensible, fully offline development platform and agent/workflow toolkit for the AI era, with an intuitive TUI, easy deployment, and complete control over your data, agents, and workflows. | 278 | 96.5 | Python | 3d |
| 8 | freestylefly/WeChatBridge 微信聊天记录一键转发到 AI Agent 与 Obsidian 的原生 macOS 工具 ai-agent · chat-history · macos · obsidian · share-extension · swift | 939 | 94.8 | Swift | 10d |
| 9 | amitshekhariitbhu/ai-system-design AI System Design - Learn how to design AI systems built on LLMs, RAG, and AI Agents step by step. ai · ai-agents · ai-engineering · ai-system · ai-system-design · ai-systems | 476 | 81.7 | Markdown | 6d |
| 10 | LockedinLabs-AI/agent-console Local-first observability for AI coding agents. Every Claude Code and Codex session's tokens, cache, models and cost, on this machine and every machine you connect through a self-hosted team hub. Presenting mode, policy hooks, Apple-signed and notarized macOS builds, attested rel ai-agents · anthropic · claude-code · codex · developer-tools · finops | 758 | 73.7 | JavaScript | 10d |
| 11 | tigerless-labs/agent-memory Long-term memory runtime for AI agents — plain Markdown as the source of truth, local ranked retrieval, and an independent sleep-time Manage layer. Claude Code and Codex share one store. No API key. agent-memory · ai-agents · claude-code · codex · llm · local-first | 1,915 | 65.8 | Python | 29d |
| 12 | jtydhr88/screenwriting-skills Professional agent skills for screenwriting, television writing and dramaturgy ai · skills | 1,468 | 59.9 | Python | 25d |
| 13 | amitshekhariitbhu/ai-engineering-course AI Engineering Course - A free and complete AI Engineering Course to learn AI Engineering step by step - from Machine Learning, Neural Networks, and Transformers to LLMs, Fine-Tuning, RAG, AI Agents, LLM Inference, Evaluation, AI Safety, and AI System Design. ai-course · ai-engineering · ai-engineering-course · ai-engineering-curriculum · large-language-models | 461 | 59.9 | Markdown | 8d |
| 14 | youngyangyang04/llm-master 大模型(LLM)全栈学习路线与中文教程🔥:覆盖 Prompt Engineering、RAG、AI Agent、MCP、微调、模型部署、Transformer、AI 编程与大厂面试,从入门到生产实践。 agent · large-language-models · llm · llm-learning · llm-tutorial · rag | 1,046 | 55.9 | — | 19d |
| 15 | aliyun/ai-agent-handbook A practical guide to building enterprise AI agents across their full lifecycle. | 1,066 | 54 | — | 20d |
| 16 | agentverse-os/AgentVerse-OS Personal cloud OS for a developer and their AI agents on a single server. One-command install on Ubuntu, then everything in the browser: a windowed desktop, isolated workspaces with VS Code, Claude Code and Codex, a store of 944 self-hosted apps, backups and updates. Access only | 984 | 53.9 | Rust | 18d |
| 17 | angel291592/Intent-Router Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before routing), the input layer for routers and typed-decision models like Jev & Laya agent-skills · agentic · ai-agents · clarification · claude-code · codex | 439 | 50 | Python | 9d |
| 18 | leter/zh-tech-writing 写中文技术文档的 Agent Skill,基于阮一峰《中文技术文档的写作规范》:短句、平实、没有 AI 腔 agent-skills · chinese · claude-code · documentation · style-guide · technical-writing | 312 | 49.6 | — | 6d |
| 19 | Jakeschincariol/linkedin-agent-skill Eleven free Claude skills that run a LinkedIn account: posts off 21 hook formulas, comments, replies, profile score, weekly plan, and a humanizer that strips the AI fingerprint and scores the draft before it goes out. agent-skills · ai-humanizer · claude · claude-code · claude-skills · content-marketing | 958 | 41.5 | Python | 23d |
| 20 | undefined-ui/second-brain-os An AI second brain that maintains itself. Full guide, starter vault, agent skills and scripts for a self-organizing knowledge base in Claude Code and Obsidian. agent-skills · ai-agents · claude · claude-code · claude-skills · knowledge-graph | 833 | 35.6 | HTML | 23d |
| 21 | malevrigns/agent-jev AgentJev-0.6B - a fast 'System One' decision model for AI Agents: feed it any unstructured state (diffs, traces, logs) and structured questions, get calibrated probability distributions back in one ~50ms forward pass. Zero output-token decoding. | 331 | 35.4 | Python | 9d |
| 22 | okf-memory/okf-agent-memory Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go. agent-memory · ai-agents · bm25 · coding-agents · developer-tools · git-native | 740 | 29.4 | Go | 25d |
| 23 | sno-ai/sno-station Sno Station — your Claude Code and Codex working as one squad on your own machine. Shared encrypted memory, agent-to-agent messaging (Reach), squad skills for handoff and cross-vendor review, and a nightly loop that rewrites the agents' own skills with your approval. Open source, agent-communication · agent-memory · agentic-ai · ai-agents · claude-code · codex | 322 | 27.4 | TypeScript | 12d |
| 24 | CosmosMind-ai/RSI-Harness RSIH — versionable, shareable agent harness: Pi coding agent + Genome config layer | 718 | 26 | TypeScript | 28d |
| 25 | i-am-manware/Manware-s-AI-Learning-Toolkit An AI toolkit that turns agents into teachers rather than code yapping machines | 650 | 25.7 | — | 25d |
| 26 | Station-Sciences/bot-crossing A video game for AI agents. Created by Jarren Rocks | 743 | 25.4 | JavaScript | 29d |
| 27 | itsmostafa/system-one-connector System One MCP connector to evaluate anything fast and cheap. Give your AI agent direct access to models like: Jev, D1, CLM and Laya mcp · mcp-tools | 337 | 24.4 | Go | 14d |
| 28 | Simreal-AI/Simreal-MLBench [Public preview] Externally scored agentic ML research benchmark: 60 tasks, real competition ground truth. Open protocol, operated evaluation. agentic · ai-evaluation · benchmark · benchmarking · calibration · llm | 219 | 22.9 | Python | 10d |
| 29 | Qiuner/QCode An explorable island world to learn AI coding and build real projects with AI agents. Powered by DeepSeek Harness. ai-agents · ai-coding · deepseek · developer-tools · gamification · godot | 347 | 20.9 | TypeScript | 17d |
| 30 | Yean-Sec/StrikeAgent_AtkBrain-Flash 由夜安团队研发的AI Agent渗透测试平台,强调自监督,自循环,自进化首创寻宝机制、三段式上升结构,涵盖红队打点、SRC、CTF,特别是在红队领域有极为亮眼的存在 | 588 | 20.5 | Python | 29d |
| 31 | haplollc/ThinkingOrbs Dotted, honestly-3D loading indicators for AI and agent interfaces in SwiftUI. Nine hand-tuned designs, two tuned sizes. ai · animation · ios · loading-indicator · spinner · swift | 241 | 20.5 | Swift | 12d |
| 32 | ZeroDayEvil/ai-security-tool 🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI. ai-hacking · ai-security-tool · cve-scanner · cybersecurity · hacking-tool · llm-security | 413 | 20 | HTML | 21d |
| 33 | kitze/skillbox Self-hosted, versioned skills library for AI agents. MCP, scoped clients, and optional Jev recommendations. | 255 | 19.2 | TypeScript | 13d |
| 34 | BoardUI/boardui React design system for agentic interfaces. Every free BoardUI component as source, with a working AI chat app on your own model key as the homepage. agentic-ai · ai-agent · ai-chat · boardui · design-system · nextjs | 551 | 19 | TypeScript | 29d |
| 35 | NiazMorshed2007/jev-review Local-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev. agent-plugin · ai-agents · claude-code · code-review · codex · coding-agents | 230 | 16.7 | TypeScript | 14d |
| 36 | Loopera-ai/loopera 面向基本面因子研究的智能体-A hypothesis-driven AI agent for fundamental factor research, with evidence-gated validation and research memory. ai · ai-agent · quant | 364 | 16.2 | — | 23d |
| 37 | Phyzicalorg/Phyzical_org Browser teleoperation data for embodied AI — elizaOS-ready episodes, trajectory_db converter, onchain provenance. The fuel station for agent robot stacks. data-platform · elizaos · embodied-ai · imitation-learning · robot-learning · robotics | 293 | 13.6 | Python | 22d |
| 38 | FankChen/tracecrate Local-first AI agent trace workbench. Inspect Claude Code, Codex and OTLP logs, compare runs, and export privacy-conscious reports. No backend or API keys. ai · ai-agents · claude-code · codex · developer-tools · local-first | 276 | 13.4 | TypeScript | 21d |
| 39 | Agenta-AI/awesome-ai-agent-platforms A curated list of open-source AI agent platforms: AI coworkers and teammates, agent builders and frameworks, workflow automation platforms, browser agents, and coding agents, with licenses. agent-frameworks · ai-agent-platforms · ai-agents · ai-coworker · awesome · awesome-list | 342 | 12.1 | Astro | 28d |
| 40 | VaderChen/YourDesk Cross-platform remote desktop for macOS and Windows with hardware acceleration, multi-monitor support, clipboard transfer for text, images, files and folders, and MCP integration for AI agents to connect to and control remote computers. clipboard · golang · macos · remote-desktop · webrtc · windows | 260 | 11.4 | Go | 23d |
| 41 | lotchuazzz-crypto/papergraph-mcp PaperGraph MCP turns math papers into evidence-grounded reading maps for AI agents: extract results, trace proof evidence, plan reading order, and review external dependencies without guessing. ai-agents · arxiv · knowledge-graph · latex · mathematics · mcp | 275 | 9.6 | Python | 29d |
| 42 | bestagentkits/design-studio-ai Open-source design workspace for AI agents and humans. Cloud editing, 3D, motion, MCP, WebMCP, CLI and BYOK. MIT. | 216 | 9.2 | TypeScript | 23d |
| 43 | LetMeHappyCode/auto-video-agent 自动流水线化生成成套「分镜插画提示词 → AI 生图 → 拼接成片」的视频,基于 Agent / Skill / Workflow 编排。 | 217 | 9.1 | Python | 24d |
| 44 | yuyuanweb/ai-picture-editor 2026 年编程导航 AI 编程实战新项目,基于 React 19 + FastAPI + LangChain + LangGraph + Konva 的 AI 修图智能体,支持自然语言驱动修图、21 种编辑工具、文生图、智能抠图、图层拆分、区域编辑、多步计划。覆盖 AI Agent 架构、SSE 实时推送、异步任务队列、Canvas 画布引擎。用一套教程掌握 AI Agent 全栈开发,不到一周学完,给你的简历增加竞争力 ai · ai-agent · ai-image · async-queue · canvas-editor · fastapi | 263 | 8.8 | Python | 30d |
| 45 | Taiquan-Zhou/RunTheRepo An AI agent for running, testing, and hardening GitHub Docker Compose apps.一个用于运行、测试和加固 GitHub Docker Compose 应用的 AI Agent。 agentic-ai · ai-agents · automated-testing · container-security · developer-tools · devops | 244 | 8.4 | Python | 29d |
- 1. feder-cr/dots1,809.2★/d
Open-source dots for the web: an AI agent with its own browser, one that does not get blocked.
- 2. zai-org/ZCode688.8★/d
Z.ai's coding agent harness. Powerful, intelligent, extensible.
- 3. nanaism/yomiyasu258★/d
AI生成された文章の不自然さを取り除き、人間にとって自然な日本語へ書き直すためのAgent Skill
A comprehensive logo-design skill for Claude, Gemini CLI, Codex and other AI agents: principles, process, SVG craft, testing tools and a 1,400+ logo reference library.
- 5. lemomo-ai/lemo-opuscar148.3★/d
43 film styles, each a reusable style prompt plus a short film made entirely in code by Claude Opus 5.5. Pick a style, bring your own story, and let your agent direct. | Opus5.5 x 43 种影片风格:风格提示词 + 纯代码样片 + 导演与技术指南
- 6. shadcn-ui/lint104.3★/d
An agent-first linter for Tailwind design systems. Write design system rules that agents can verify.
- 7. openJiuwen-ai/iCode96.5★/d
A lightweight, extensible, fully offline development platform and agent/workflow toolkit for the AI era, with an intuitive TUI, easy deployment, and complete control over your data, agents, and workflows.
- 8. freestylefly/WeChatBridge94.8★/d
微信聊天记录一键转发到 AI Agent 与 Obsidian 的原生 macOS 工具
AI System Design - Learn how to design AI systems built on LLMs, RAG, and AI Agents step by step.
Local-first observability for AI coding agents. Every Claude Code and Codex session's tokens, cache, models and cost, on this machine and every machine you connect through a self-hosted team hub. Presenting mode, policy hooks, Apple-signed and notarized macOS builds, attested rel
Long-term memory runtime for AI agents — plain Markdown as the source of truth, local ranked retrieval, and an independent sleep-time Manage layer. Claude Code and Codex share one store. No API key.
Professional agent skills for screenwriting, television writing and dramaturgy
AI Engineering Course - A free and complete AI Engineering Course to learn AI Engineering step by step - from Machine Learning, Neural Networks, and Transformers to LLMs, Fine-Tuning, RAG, AI Agents, LLM Inference, Evaluation, AI Safety, and AI System Design.
大模型(LLM)全栈学习路线与中文教程🔥:覆盖 Prompt Engineering、RAG、AI Agent、MCP、微调、模型部署、Transformer、AI 编程与大厂面试,从入门到生产实践。
A practical guide to building enterprise AI agents across their full lifecycle.
Personal cloud OS for a developer and their AI agents on a single server. One-command install on Ubuntu, then everything in the browser: a windowed desktop, isolated workspaces with VS Code, Claude Code and Codex, a store of 944 self-hosted apps, backups and updates. Access only
Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before routing), the input layer for routers and typed-decision models like Jev & Laya
- 18. leter/zh-tech-writing49.6★/d
写中文技术文档的 Agent Skill,基于阮一峰《中文技术文档的写作规范》:短句、平实、没有 AI 腔
Eleven free Claude skills that run a LinkedIn account: posts off 21 hook formulas, comments, replies, profile score, weekly plan, and a humanizer that strips the AI fingerprint and scores the draft before it goes out.
An AI second brain that maintains itself. Full guide, starter vault, agent skills and scripts for a self-organizing knowledge base in Claude Code and Obsidian.
- 21. malevrigns/agent-jev35.4★/d
AgentJev-0.6B - a fast 'System One' decision model for AI Agents: feed it any unstructured state (diffs, traces, logs) and structured questions, get calibrated probability distributions back in one ~50ms forward pass. Zero output-token decoding.
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.
- 23. sno-ai/sno-station27.4★/d
Sno Station — your Claude Code and Codex working as one squad on your own machine. Shared encrypted memory, agent-to-agent messaging (Reach), squad skills for handoff and cross-vendor review, and a nightly loop that rewrites the agents' own skills with your approval. Open source,
RSIH — versionable, shareable agent harness: Pi coding agent + Genome config layer
An AI toolkit that turns agents into teachers rather than code yapping machines
A video game for AI agents. Created by Jarren Rocks
System One MCP connector to evaluate anything fast and cheap. Give your AI agent direct access to models like: Jev, D1, CLM and Laya
[Public preview] Externally scored agentic ML research benchmark: 60 tasks, real competition ground truth. Open protocol, operated evaluation.
- 29. Qiuner/QCode20.9★/d
An explorable island world to learn AI coding and build real projects with AI agents. Powered by DeepSeek Harness.
由夜安团队研发的AI Agent渗透测试平台,强调自监督,自循环,自进化首创寻宝机制、三段式上升结构,涵盖红队打点、SRC、CTF,特别是在红队领域有极为亮眼的存在
- 31. haplollc/ThinkingOrbs20.5★/d
Dotted, honestly-3D loading indicators for AI and agent interfaces in SwiftUI. Nine hand-tuned designs, two tuned sizes.
🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI.
- 33. kitze/skillbox19.2★/d
Self-hosted, versioned skills library for AI agents. MCP, scoped clients, and optional Jev recommendations.
- 34. BoardUI/boardui19★/d
React design system for agentic interfaces. Every free BoardUI component as source, with a working AI chat app on your own model key as the homepage.
Local-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev.
- 36. Loopera-ai/loopera16.2★/d
面向基本面因子研究的智能体-A hypothesis-driven AI agent for fundamental factor research, with evidence-gated validation and research memory.
- 37. Phyzicalorg/Phyzical_org13.6★/d
Browser teleoperation data for embodied AI — elizaOS-ready episodes, trajectory_db converter, onchain provenance. The fuel station for agent robot stacks.
- 38. FankChen/tracecrate13.4★/d
Local-first AI agent trace workbench. Inspect Claude Code, Codex and OTLP logs, compare runs, and export privacy-conscious reports. No backend or API keys.
A curated list of open-source AI agent platforms: AI coworkers and teammates, agent builders and frameworks, workflow automation platforms, browser agents, and coding agents, with licenses.
- 40. VaderChen/YourDesk11.4★/d
Cross-platform remote desktop for macOS and Windows with hardware acceleration, multi-monitor support, clipboard transfer for text, images, files and folders, and MCP integration for AI agents to connect to and control remote computers.
PaperGraph MCP turns math papers into evidence-grounded reading maps for AI agents: extract results, trace proof evidence, plan reading order, and review external dependencies without guessing.
Open-source design workspace for AI agents and humans. Cloud editing, 3D, motion, MCP, WebMCP, CLI and BYOK. MIT.
自动流水线化生成成套「分镜插画提示词 → AI 生图 → 拼接成片」的视频,基于 Agent / Skill / Workflow 编排。
2026 年编程导航 AI 编程实战新项目,基于 React 19 + FastAPI + LangChain + LangGraph + Konva 的 AI 修图智能体,支持自然语言驱动修图、21 种编辑工具、文生图、智能抠图、图层拆分、区域编辑、多步计划。覆盖 AI Agent 架构、SSE 实时推送、异步任务队列、Canvas 画布引擎。用一套教程掌握 AI Agent 全栈开发,不到一周学完,给你的简历增加竞争力
An AI agent for running, testing, and hardening GitHub Docker Compose apps.一个用于运行、测试和加固 GitHub Docker Compose 应用的 AI Agent。
$99/month · Cancel anytime · Every Tool included
Same input. Real output.
This is one real run of GitHub Discovery on October 1, 2026, exactly as it came out. The data is live, so yours will be different.
Without it
- Guess at GitHub search queries
- Open repos one by one to check their age and stars
- Work out stars-per-day in a spreadsheet
- Repeat it next week
With GitHub Discovery: what went in
- Created in the last (days)
- 30
- Minimum stars
- 200
- Keyword
- ai agent
- Sort by
- Star velocity (stars/day)
What came out
| # | Repo | Stars | ★/day | Lang | Age |
|---|---|---|---|---|---|
| 1 | feder-cr/dots Open-source dots for the web: an AI agent with its own browser, one that does not get blocked. ai-agent · ai-agents · ai-browser · anti-detect-browser · browser-agent · browser-automation | 1,933 | 1,809.2 | Python | 1d |
| 2 | zai-org/ZCode Z.ai's coding agent harness. Powerful, intelligent, extensible. | 7,253 | 688.8 | TypeScript | 11d |
| 3 | nanaism/yomiyasu AI生成された文章の不自然さを取り除き、人間にとって自然な日本語へ書き直すためのAgent Skill agent-skills · ai-writing · claude-code · codex · cursor · japanese | 258 | 258 | Python | 1d |
| 4 | kaankiziltug/logo-design-skill A comprehensive logo-design skill for Claude, Gemini CLI, Codex and other AI agents: principles, process, SVG craft, testing tools and a 1,400+ logo reference library. agent-skills · branding · claude · claude-skills · codex · gemini-cli | 1,110 | 237.1 | HTML | 5d |
| 5 | lemomo-ai/lemo-opuscar 43 film styles, each a reusable style prompt plus a short film made entirely in code by Claude Opus 5.5. Pick a style, bring your own story, and let your agent direct. | Opus5.5 x 43 种影片风格:风格提示词 + 纯代码样片 + 导演与技术指南 ai-agents · ai-video · animation · canvas · claude · claude-code | 683 | 148.3 | JavaScript | 5d |
| 6 | shadcn-ui/lint An agent-first linter for Tailwind design systems. Write design system rules that agents can verify. agents · ai · design · design-system · design-tools · shadcn | 2,976 | 104.3 | TypeScript | 29d |
| 7 | openJiuwen-ai/iCode A lightweight, extensible, fully offline development platform and agent/workflow toolkit for the AI era, with an intuitive TUI, easy deployment, and complete control over your data, agents, and workflows. | 278 | 96.5 | Python | 3d |
| 8 | freestylefly/WeChatBridge 微信聊天记录一键转发到 AI Agent 与 Obsidian 的原生 macOS 工具 ai-agent · chat-history · macos · obsidian · share-extension · swift | 939 | 94.8 | Swift | 10d |
| 9 | amitshekhariitbhu/ai-system-design AI System Design - Learn how to design AI systems built on LLMs, RAG, and AI Agents step by step. ai · ai-agents · ai-engineering · ai-system · ai-system-design · ai-systems | 476 | 81.7 | Markdown | 6d |
| 10 | LockedinLabs-AI/agent-console Local-first observability for AI coding agents. Every Claude Code and Codex session's tokens, cache, models and cost, on this machine and every machine you connect through a self-hosted team hub. Presenting mode, policy hooks, Apple-signed and notarized macOS builds, attested rel ai-agents · anthropic · claude-code · codex · developer-tools · finops | 758 | 73.7 | JavaScript | 10d |
| 11 | tigerless-labs/agent-memory Long-term memory runtime for AI agents — plain Markdown as the source of truth, local ranked retrieval, and an independent sleep-time Manage layer. Claude Code and Codex share one store. No API key. agent-memory · ai-agents · claude-code · codex · llm · local-first | 1,915 | 65.8 | Python | 29d |
| 12 | jtydhr88/screenwriting-skills Professional agent skills for screenwriting, television writing and dramaturgy ai · skills | 1,468 | 59.9 | Python | 25d |
| 13 | amitshekhariitbhu/ai-engineering-course AI Engineering Course - A free and complete AI Engineering Course to learn AI Engineering step by step - from Machine Learning, Neural Networks, and Transformers to LLMs, Fine-Tuning, RAG, AI Agents, LLM Inference, Evaluation, AI Safety, and AI System Design. ai-course · ai-engineering · ai-engineering-course · ai-engineering-curriculum · large-language-models | 461 | 59.9 | Markdown | 8d |
| 14 | youngyangyang04/llm-master 大模型(LLM)全栈学习路线与中文教程🔥:覆盖 Prompt Engineering、RAG、AI Agent、MCP、微调、模型部署、Transformer、AI 编程与大厂面试,从入门到生产实践。 agent · large-language-models · llm · llm-learning · llm-tutorial · rag | 1,046 | 55.9 | — | 19d |
| 15 | aliyun/ai-agent-handbook A practical guide to building enterprise AI agents across their full lifecycle. | 1,066 | 54 | — | 20d |
| 16 | agentverse-os/AgentVerse-OS Personal cloud OS for a developer and their AI agents on a single server. One-command install on Ubuntu, then everything in the browser: a windowed desktop, isolated workspaces with VS Code, Claude Code and Codex, a store of 944 self-hosted apps, backups and updates. Access only | 984 | 53.9 | Rust | 18d |
| 17 | angel291592/Intent-Router Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before routing), the input layer for routers and typed-decision models like Jev & Laya agent-skills · agentic · ai-agents · clarification · claude-code · codex | 439 | 50 | Python | 9d |
| 18 | leter/zh-tech-writing 写中文技术文档的 Agent Skill,基于阮一峰《中文技术文档的写作规范》:短句、平实、没有 AI 腔 agent-skills · chinese · claude-code · documentation · style-guide · technical-writing | 312 | 49.6 | — | 6d |
| 19 | Jakeschincariol/linkedin-agent-skill Eleven free Claude skills that run a LinkedIn account: posts off 21 hook formulas, comments, replies, profile score, weekly plan, and a humanizer that strips the AI fingerprint and scores the draft before it goes out. agent-skills · ai-humanizer · claude · claude-code · claude-skills · content-marketing | 958 | 41.5 | Python | 23d |
| 20 | undefined-ui/second-brain-os An AI second brain that maintains itself. Full guide, starter vault, agent skills and scripts for a self-organizing knowledge base in Claude Code and Obsidian. agent-skills · ai-agents · claude · claude-code · claude-skills · knowledge-graph | 833 | 35.6 | HTML | 23d |
| 21 | malevrigns/agent-jev AgentJev-0.6B - a fast 'System One' decision model for AI Agents: feed it any unstructured state (diffs, traces, logs) and structured questions, get calibrated probability distributions back in one ~50ms forward pass. Zero output-token decoding. | 331 | 35.4 | Python | 9d |
| 22 | okf-memory/okf-agent-memory Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go. agent-memory · ai-agents · bm25 · coding-agents · developer-tools · git-native | 740 | 29.4 | Go | 25d |
| 23 | sno-ai/sno-station Sno Station — your Claude Code and Codex working as one squad on your own machine. Shared encrypted memory, agent-to-agent messaging (Reach), squad skills for handoff and cross-vendor review, and a nightly loop that rewrites the agents' own skills with your approval. Open source, agent-communication · agent-memory · agentic-ai · ai-agents · claude-code · codex | 322 | 27.4 | TypeScript | 12d |
| 24 | CosmosMind-ai/RSI-Harness RSIH — versionable, shareable agent harness: Pi coding agent + Genome config layer | 718 | 26 | TypeScript | 28d |
| 25 | i-am-manware/Manware-s-AI-Learning-Toolkit An AI toolkit that turns agents into teachers rather than code yapping machines | 650 | 25.7 | — | 25d |
| 26 | Station-Sciences/bot-crossing A video game for AI agents. Created by Jarren Rocks | 743 | 25.4 | JavaScript | 29d |
| 27 | itsmostafa/system-one-connector System One MCP connector to evaluate anything fast and cheap. Give your AI agent direct access to models like: Jev, D1, CLM and Laya mcp · mcp-tools | 337 | 24.4 | Go | 14d |
| 28 | Simreal-AI/Simreal-MLBench [Public preview] Externally scored agentic ML research benchmark: 60 tasks, real competition ground truth. Open protocol, operated evaluation. agentic · ai-evaluation · benchmark · benchmarking · calibration · llm | 219 | 22.9 | Python | 10d |
| 29 | Qiuner/QCode An explorable island world to learn AI coding and build real projects with AI agents. Powered by DeepSeek Harness. ai-agents · ai-coding · deepseek · developer-tools · gamification · godot | 347 | 20.9 | TypeScript | 17d |
| 30 | Yean-Sec/StrikeAgent_AtkBrain-Flash 由夜安团队研发的AI Agent渗透测试平台,强调自监督,自循环,自进化首创寻宝机制、三段式上升结构,涵盖红队打点、SRC、CTF,特别是在红队领域有极为亮眼的存在 | 588 | 20.5 | Python | 29d |
| 31 | haplollc/ThinkingOrbs Dotted, honestly-3D loading indicators for AI and agent interfaces in SwiftUI. Nine hand-tuned designs, two tuned sizes. ai · animation · ios · loading-indicator · spinner · swift | 241 | 20.5 | Swift | 12d |
| 32 | ZeroDayEvil/ai-security-tool 🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI. ai-hacking · ai-security-tool · cve-scanner · cybersecurity · hacking-tool · llm-security | 413 | 20 | HTML | 21d |
| 33 | kitze/skillbox Self-hosted, versioned skills library for AI agents. MCP, scoped clients, and optional Jev recommendations. | 255 | 19.2 | TypeScript | 13d |
| 34 | BoardUI/boardui React design system for agentic interfaces. Every free BoardUI component as source, with a working AI chat app on your own model key as the homepage. agentic-ai · ai-agent · ai-chat · boardui · design-system · nextjs | 551 | 19 | TypeScript | 29d |
| 35 | NiazMorshed2007/jev-review Local-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev. agent-plugin · ai-agents · claude-code · code-review · codex · coding-agents | 230 | 16.7 | TypeScript | 14d |
| 36 | Loopera-ai/loopera 面向基本面因子研究的智能体-A hypothesis-driven AI agent for fundamental factor research, with evidence-gated validation and research memory. ai · ai-agent · quant | 364 | 16.2 | — | 23d |
| 37 | Phyzicalorg/Phyzical_org Browser teleoperation data for embodied AI — elizaOS-ready episodes, trajectory_db converter, onchain provenance. The fuel station for agent robot stacks. data-platform · elizaos · embodied-ai · imitation-learning · robot-learning · robotics | 293 | 13.6 | Python | 22d |
| 38 | FankChen/tracecrate Local-first AI agent trace workbench. Inspect Claude Code, Codex and OTLP logs, compare runs, and export privacy-conscious reports. No backend or API keys. ai · ai-agents · claude-code · codex · developer-tools · local-first | 276 | 13.4 | TypeScript | 21d |
| 39 | Agenta-AI/awesome-ai-agent-platforms A curated list of open-source AI agent platforms: AI coworkers and teammates, agent builders and frameworks, workflow automation platforms, browser agents, and coding agents, with licenses. agent-frameworks · ai-agent-platforms · ai-agents · ai-coworker · awesome · awesome-list | 342 | 12.1 | Astro | 28d |
| 40 | VaderChen/YourDesk Cross-platform remote desktop for macOS and Windows with hardware acceleration, multi-monitor support, clipboard transfer for text, images, files and folders, and MCP integration for AI agents to connect to and control remote computers. clipboard · golang · macos · remote-desktop · webrtc · windows | 260 | 11.4 | Go | 23d |
| 41 | lotchuazzz-crypto/papergraph-mcp PaperGraph MCP turns math papers into evidence-grounded reading maps for AI agents: extract results, trace proof evidence, plan reading order, and review external dependencies without guessing. ai-agents · arxiv · knowledge-graph · latex · mathematics · mcp | 275 | 9.6 | Python | 29d |
| 42 | bestagentkits/design-studio-ai Open-source design workspace for AI agents and humans. Cloud editing, 3D, motion, MCP, WebMCP, CLI and BYOK. MIT. | 216 | 9.2 | TypeScript | 23d |
| 43 | LetMeHappyCode/auto-video-agent 自动流水线化生成成套「分镜插画提示词 → AI 生图 → 拼接成片」的视频,基于 Agent / Skill / Workflow 编排。 | 217 | 9.1 | Python | 24d |
| 44 | yuyuanweb/ai-picture-editor 2026 年编程导航 AI 编程实战新项目,基于 React 19 + FastAPI + LangChain + LangGraph + Konva 的 AI 修图智能体,支持自然语言驱动修图、21 种编辑工具、文生图、智能抠图、图层拆分、区域编辑、多步计划。覆盖 AI Agent 架构、SSE 实时推送、异步任务队列、Canvas 画布引擎。用一套教程掌握 AI Agent 全栈开发,不到一周学完,给你的简历增加竞争力 ai · ai-agent · ai-image · async-queue · canvas-editor · fastapi | 263 | 8.8 | Python | 30d |
| 45 | Taiquan-Zhou/RunTheRepo An AI agent for running, testing, and hardening GitHub Docker Compose apps.一个用于运行、测试和加固 GitHub Docker Compose 应用的 AI Agent。 agentic-ai · ai-agents · automated-testing · container-security · developer-tools · devops | 244 | 8.4 | Python | 29d |
- 1. feder-cr/dots1,809.2★/d
Open-source dots for the web: an AI agent with its own browser, one that does not get blocked.
- 2. zai-org/ZCode688.8★/d
Z.ai's coding agent harness. Powerful, intelligent, extensible.
- 3. nanaism/yomiyasu258★/d
AI生成された文章の不自然さを取り除き、人間にとって自然な日本語へ書き直すためのAgent Skill
A comprehensive logo-design skill for Claude, Gemini CLI, Codex and other AI agents: principles, process, SVG craft, testing tools and a 1,400+ logo reference library.
- 5. lemomo-ai/lemo-opuscar148.3★/d
43 film styles, each a reusable style prompt plus a short film made entirely in code by Claude Opus 5.5. Pick a style, bring your own story, and let your agent direct. | Opus5.5 x 43 种影片风格:风格提示词 + 纯代码样片 + 导演与技术指南
- 6. shadcn-ui/lint104.3★/d
An agent-first linter for Tailwind design systems. Write design system rules that agents can verify.
- 7. openJiuwen-ai/iCode96.5★/d
A lightweight, extensible, fully offline development platform and agent/workflow toolkit for the AI era, with an intuitive TUI, easy deployment, and complete control over your data, agents, and workflows.
- 8. freestylefly/WeChatBridge94.8★/d
微信聊天记录一键转发到 AI Agent 与 Obsidian 的原生 macOS 工具
AI System Design - Learn how to design AI systems built on LLMs, RAG, and AI Agents step by step.
Local-first observability for AI coding agents. Every Claude Code and Codex session's tokens, cache, models and cost, on this machine and every machine you connect through a self-hosted team hub. Presenting mode, policy hooks, Apple-signed and notarized macOS builds, attested rel
Long-term memory runtime for AI agents — plain Markdown as the source of truth, local ranked retrieval, and an independent sleep-time Manage layer. Claude Code and Codex share one store. No API key.
Professional agent skills for screenwriting, television writing and dramaturgy
AI Engineering Course - A free and complete AI Engineering Course to learn AI Engineering step by step - from Machine Learning, Neural Networks, and Transformers to LLMs, Fine-Tuning, RAG, AI Agents, LLM Inference, Evaluation, AI Safety, and AI System Design.
大模型(LLM)全栈学习路线与中文教程🔥:覆盖 Prompt Engineering、RAG、AI Agent、MCP、微调、模型部署、Transformer、AI 编程与大厂面试,从入门到生产实践。
A practical guide to building enterprise AI agents across their full lifecycle.
Personal cloud OS for a developer and their AI agents on a single server. One-command install on Ubuntu, then everything in the browser: a windowed desktop, isolated workspaces with VS Code, Claude Code and Codex, a store of 944 self-hosted apps, backups and updates. Access only
Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before routing), the input layer for routers and typed-decision models like Jev & Laya
- 18. leter/zh-tech-writing49.6★/d
写中文技术文档的 Agent Skill,基于阮一峰《中文技术文档的写作规范》:短句、平实、没有 AI 腔
Eleven free Claude skills that run a LinkedIn account: posts off 21 hook formulas, comments, replies, profile score, weekly plan, and a humanizer that strips the AI fingerprint and scores the draft before it goes out.
An AI second brain that maintains itself. Full guide, starter vault, agent skills and scripts for a self-organizing knowledge base in Claude Code and Obsidian.
- 21. malevrigns/agent-jev35.4★/d
AgentJev-0.6B - a fast 'System One' decision model for AI Agents: feed it any unstructured state (diffs, traces, logs) and structured questions, get calibrated probability distributions back in one ~50ms forward pass. Zero output-token decoding.
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.
- 23. sno-ai/sno-station27.4★/d
Sno Station — your Claude Code and Codex working as one squad on your own machine. Shared encrypted memory, agent-to-agent messaging (Reach), squad skills for handoff and cross-vendor review, and a nightly loop that rewrites the agents' own skills with your approval. Open source,
RSIH — versionable, shareable agent harness: Pi coding agent + Genome config layer
An AI toolkit that turns agents into teachers rather than code yapping machines
A video game for AI agents. Created by Jarren Rocks
System One MCP connector to evaluate anything fast and cheap. Give your AI agent direct access to models like: Jev, D1, CLM and Laya
[Public preview] Externally scored agentic ML research benchmark: 60 tasks, real competition ground truth. Open protocol, operated evaluation.
- 29. Qiuner/QCode20.9★/d
An explorable island world to learn AI coding and build real projects with AI agents. Powered by DeepSeek Harness.
由夜安团队研发的AI Agent渗透测试平台,强调自监督,自循环,自进化首创寻宝机制、三段式上升结构,涵盖红队打点、SRC、CTF,特别是在红队领域有极为亮眼的存在
- 31. haplollc/ThinkingOrbs20.5★/d
Dotted, honestly-3D loading indicators for AI and agent interfaces in SwiftUI. Nine hand-tuned designs, two tuned sizes.
🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI.
- 33. kitze/skillbox19.2★/d
Self-hosted, versioned skills library for AI agents. MCP, scoped clients, and optional Jev recommendations.
- 34. BoardUI/boardui19★/d
React design system for agentic interfaces. Every free BoardUI component as source, with a working AI chat app on your own model key as the homepage.
Local-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev.
- 36. Loopera-ai/loopera16.2★/d
面向基本面因子研究的智能体-A hypothesis-driven AI agent for fundamental factor research, with evidence-gated validation and research memory.
- 37. Phyzicalorg/Phyzical_org13.6★/d
Browser teleoperation data for embodied AI — elizaOS-ready episodes, trajectory_db converter, onchain provenance. The fuel station for agent robot stacks.
- 38. FankChen/tracecrate13.4★/d
Local-first AI agent trace workbench. Inspect Claude Code, Codex and OTLP logs, compare runs, and export privacy-conscious reports. No backend or API keys.
A curated list of open-source AI agent platforms: AI coworkers and teammates, agent builders and frameworks, workflow automation platforms, browser agents, and coding agents, with licenses.
- 40. VaderChen/YourDesk11.4★/d
Cross-platform remote desktop for macOS and Windows with hardware acceleration, multi-monitor support, clipboard transfer for text, images, files and folders, and MCP integration for AI agents to connect to and control remote computers.
PaperGraph MCP turns math papers into evidence-grounded reading maps for AI agents: extract results, trace proof evidence, plan reading order, and review external dependencies without guessing.
Open-source design workspace for AI agents and humans. Cloud editing, 3D, motion, MCP, WebMCP, CLI and BYOK. MIT.
自动流水线化生成成套「分镜插画提示词 → AI 生图 → 拼接成片」的视频,基于 Agent / Skill / Workflow 编排。
2026 年编程导航 AI 编程实战新项目,基于 React 19 + FastAPI + LangChain + LangGraph + Konva 的 AI 修图智能体,支持自然语言驱动修图、21 种编辑工具、文生图、智能抠图、图层拆分、区域编辑、多步计划。覆盖 AI Agent 架构、SSE 实时推送、异步任务队列、Canvas 画布引擎。用一套教程掌握 AI Agent 全栈开发,不到一周学完,给你的简历增加竞争力
An AI agent for running, testing, and hardening GitHub Docker Compose apps.一个用于运行、测试和加固 GitHub Docker Compose 应用的 AI Agent。
- 45
- repos ranked in this run
- 1,809
- stars a day: the fastest repo it found
- 1.1s
- from Run to results
How it works
- 1
Set your window
Pick how new (1 to 90 days), a minimum star count, and optionally a language, topic or keyword.
- 2
It ranks by velocity
Every matching public repo, sorted by stars per day since it was created. New and fast beats old and big.
- 3
Take the list
Up to 50 repos with description, stars, stars per day, language, age and topics. Export to CSV, Markdown or JSON.
Who it's for
Creators covering AI
You need something new to post before the same five accounts cover it.
Builders
You want an open-source project to build on, integrate or turn into a product.
Agencies and newsletter writers
You owe clients or readers a what's-new list every week.
Prefer to learn the method by hand first? Read the free guide: Find GitHub Repos Before They Go Viral.
What it does
- Star-velocity ranking (stars per day since creation)
- Filter by age, minimum stars, language, topic or keyword
- Sort by velocity, total stars or newest
- Export to CSV, Markdown or JSON
- Saved runs you can star and reopen
What you need
- Runs in your browser on phone, tablet or computer. Nothing to install.
- No API keys or extra subscriptions: usage is included in the membership.
- No GitHub account needed. If GitHub is rate-limiting searches, a run asks you to try again in a minute.
- Up to 60 runs a day.
GitHub Discovery is included in The AI Stack
$99 a month, billed monthly until you cancel. No separate charge for GitHub Discovery, no usage fees.
Also in your membership
- 4 more Tools: Trend Radar, Prompt Builder, Script Generator, Page Scraper
- Every Blueprint: step-by-step systems to help you build
- Selected source code and components (first releases coming soon)
- The member community: ask questions, get help, share progress
By joining you agree to the Terms and Privacy Policy. Cancel any time. Refund Policy.
Already a member? Log in and you'll go straight to GitHub Discovery.
Ready-to-Launch Businesses are not part of the membership. They're separate and by application.
Questions
Do I need a GitHub account or token?
No. The search runs on our servers against GitHub's public search.
How fresh is the data?
Live. Each run queries GitHub at that moment, so the list changes day to day.
Can I use the results in my content or products?
Yes. It's public repository data. Check each project's own licence before you build on its code.
What does it cost?
GitHub Discovery is part of The AI Stack membership: $99 a month, billed monthly until you cancel. There's no separate charge for this Tool and no usage fees on top.
Is there a limit?
Up to 60 runs a day, which resets daily.
What happens after I pay?
Checkout creates your account with the email you pay with and takes you straight into GitHub Discovery. If the payment takes a moment to confirm, you'll see that and can refresh safely: you're never charged twice.
Can I cancel?
Yes, any time from your billing settings. You keep access until the end of the month you've paid for. Our Refund Policy explains refunds.
Do I get the source code?
Using GitHub Discovery is included. The membership also includes selected source code and components under our licence (see License). Ready-to-Launch Businesses are separate and by application.
Where do I get help?
Ask in the member community, which is included, or email support@jointheaistack.com.
Policies: Refund Policy, Terms, Privacy, License.
Find the next viral AI repo before everyone posts about it.
Get GitHub Discovery$99/month · Cancel anytime · Lands you straight in GitHub Discovery