Rank 1
A macOS menu bar app that safely keeps your MacBook awake with the lid closed, for timed agent sessions.
Breakout week: +1.3k this week vs ~2
+1.3k stars this week
#ai-agents
Emerging AI agents repos
Small repositories, under 1,000 stars a week ago, gaining stars at least twice as fast as in a usual week. They already had a history to speed up from: these are not this week’s new repos. How emerging works
Showing 58 of 58 · stars this week against the usual week · min. 20
Rank 2
Autonomous AI sales agent powered by Claude Code or Codex. Finds prospects, writes cold emails, sends campaigns, handles replies, books meetings. No API keys needed — runs on your Claude or Codex subscriptions.
Breakout week: +1k this week vs ~5
+1k stars this week
Rank 3
The operational superset of Pi Coding Agent — everything Pi, plus observability, governance, recovery, evaluation and multi-agent orchestration. Pi Coding Agent 的运维超集——完整继承 Pi 的核心能力，并进一步扩展可观测性、治理、故障恢复、评估和多 Agent 编排。
58× its usual pace: +925 this week vs ~16
+925 stars this week
Rank 4
For You Agent——AI 时代的个人随身数字人格。把你的模型、AI 账号、技能、提示词和工作方式，带到每一个 AI 工具里。
56× its usual pace: +620 this week vs ~11
+620 stars this week
Rank 5
Local-first AI teammate workspace. Bring your own models and keep project conversations, tool approvals, and results together.
+555 stars this week after quiet weeks
+555 stars this week
Rank 6
+570 stars this week
Rank 7
Local-first AI agent workspace for multi-agent collaboration, agent orchestration, scoped permissions, evidence-aware runs, and human-in-the-loop decisions.
49× its usual pace: +488 this week vs ~10
+488 stars this week
Rank 8
Scenario for Blender: AI images, video, 3D and PBR materials in the viewport, plus a local MCP server for agents. Blender 5.0+ extension, GPL-3.0-or-later.
Breakout week: +461 this week vs ~3
+461 stars this week
+1 more from scenario-labs
Rank 28
Get production-ready images, video, audio, and 3D from any AI agent: skills that pick the right model, price before spending, and keep characters and brands consistent through the Scenario MCP, plus expert teams that drive Blender, Maya, ZBrush, Unreal, and Unity.
16× its usual pace: +82 this week vs ~5
+82 stars this week
Rank 9
YouTube transcript skills for Claude Code, Cursor and Codex. Agent Skills to get transcripts, search YouTube, read channels and playlists, and bulk export.
+250 stars this week after quiet weeks
+250 stars this week
Rank 10
A local multi-agent harness that works with your existing Claude Code, Codex subscriptions, allows you to run an office of agents
25× its usual pace: +394 this week vs ~16
+394 stars this week
Rank 11
Self-hosted AI trading terminal for Hyperliquid: chart read in, validated Freqtrade/Nautilus-Trader strategy out, deployed live.
49× its usual pace: +244 this week vs ~5
+244 stars this week
Rank 12
High-performance native SSH client in Rust, GPU-rendered with GPUI: host manager, terminal, WinSCP-style SFTP and port forwarding. Rust 编写、GPU 渲染的原生 SSH 客户端。
+191 stars this week after quiet weeks
+191 stars this week
Rank 13
Autonomous GitHub Project build → QA → review loop for Claude Code. One-line install.
Breakout week: +185 this week vs ~1
+185 stars this week
Rank 14
Claude Code skill for Dribbble-level product launch videos from code. One shape that never cuts, springs, a cursor that drives every change, beat-synced SFX, 60 fps motion blur. v2: use with Claude Opus 5.5.
34× its usual pace: +170 this week vs ~5
+170 stars this week
Rank 15
响应快 省Token， Let AI agents (Claude Code, Codex, any MCP client) drive a real iPhone · 让 AI agent 操作真实 iPhone：读屏、点击、输入、滑动，USB 或 Wi-Fi，自有 XCTest runner，可回放的 App flow。Open source, self-hosted on macOS.
29× its usual pace: +145 this week vs ~5
+145 stars this week
Rank 16
Operant Beta：以多模型分工、按需通信、分层审批和经验复用为核心设计，在本机组织可追溯的 Agent 协作，并通过模块化架构扩展能力。
Breakout week: +145 this week vs ~1
+145 stars this week
Rank 17
24 open-source agent skills that turn taste into instructions — ten genre identities, each proven by a same-brief before/after demo
+142 stars this week after quiet weeks
+142 stars this week
Rank 18
+250 stars this week
Rank 19
Cayu is an open-source Python framework for building and running domain-specific long-horizon agents, with a durable runtime built in.
28× its usual pace: +111 this week vs ~4
+111 stars this week
Rank 20
让 AI 编程助手复用本机工具并按项目要求选对版本；Windows 工具发现、验证与项目规则接入。
+111 stars this week after quiet weeks
+111 stars this week
Rank 21
An evaluation harness for medical/health AI agents — reproduce and cover multiple benchmarks under one scoring discipline
54× its usual pace: +107 this week vs ~2
+107 stars this week
Rank 22
从 Vibe Coding 到 Vibe World：AI原生的可编程三维世界，面向元宇宙、Science、世界模型的开源底座
+102 stars this week after quiet weeks
+102 stars this week
Rank 23
Text-to-SQL access control: schema selection that gives the LLM only the tables the caller is allowed to read. Row-level-security-aware table pruning for LangChain, MCP and any SQL agent. Postgres, Oracle, MySQL, SQL Server. 75% fewer prompt tokens.
96× its usual pace: +96 this week vs ~1
+96 stars this week
Rank 24
A hierarchical memory framework for personalized presentation agents. Try it at memslides.com.
23× its usual pace: +91 this week vs ~4
+91 stars this week
Rank 25
中国 A 股数据基础设施。55个日更数据集：股票行情、基本面、期货数据、资金面、公告事件、指数行业、宏观与风险。行级溯源、PIT 语义、复权与历史成分内置，MCP 原生。自托管，零注册、零 API Token
9× its usual pace: +234 this week vs ~26
+234 stars this week
Rank 26
Turn papers and research topics into source-linked commercialization assessment drafts. Python, FastAPI and CrewAI; auditable scoring, checkpoint recovery and durable request receipts.
9× its usual pace: +135 this week vs ~15
+135 stars this week
Rank 27
Pi — A cross-platform AI coding agent, bringing the Claude Code experience to your desktop. No environment setup, no terminal commands. Download and start coding right away.
9× its usual pace: +304 this week vs ~34
+304 stars this week
Rank 29
Auditable AI-assisted scientific research: evidence-gated hypothesis routing, isolated computation, independent replication, and anomaly-sensitive representation repair.
+81 stars this week after quiet weeks
+81 stars this week
Rank 30
An elegant, information-rich status, menu, and side bar for Pi Agent.
13× its usual pace: +77 this week vs ~6
+77 stars this week
Rank 31
A kanban board for coordinating multiple AI coding agents, with isolated git worktrees and independent reviewer agents.
9× its usual pace: +75 this week vs ~8
+75 stars this week
Rank 32
Cosmonapse is an open-source event-driven runtime for distributed AI agents. Build agent systems around typed signals, asynchronous messaging, and independent services instead of a centralized orchestrator.
+73 stars this week after quiet weeks
+73 stars this week
Rank 33
Open-source AI security verification for model artifacts, live endpoints, MCP servers, and recorded agent traces. Reproducible evidence for release decisions.
24× its usual pace: +73 this week vs ~3
+73 stars this week
Rank 34
CLI tool & MCP for Indonesian stock market (IDX) analysis. Built for humans and AI agents.
+71 stars this week after quiet weeks
+71 stars this week
Rank 35
Your coding agent paints pixel art in the Aseprite window you already have open. MCP server + skills for Claude Code, omp, Codex, Gemini CLI and Cursor — draws, looks at its work, animates and exports. Gallery and benchmark: pixeli.pebbly.space
6× its usual pace: +146 this week vs ~23
+146 stars this week
Rank 36
Open-source, self-hostable alternative to ChatGPT and NotebookLM — agents, skills, MCP, and cited knowledge in one AI workstation. Your models, your infrastructure. Free to use GPT 6: https://SourceWeft.com/
+63 stars this week after quiet weeks
+63 stars this week
Rank 37
+63 stars this week
Rank 38
同行 · Together — 面向自主经营、任务与全成就推进的 Stardew Valley AI Agent。原生工具、持续计划、游戏记忆与玩家辅助。
+62 stars this week after quiet weeks
+62 stars this week
Rank 39
The phone is your last API. Give your agent a voice to call any US business - reservations, inquiries, quotes. Verified outcome + transcript. First 250 calls free.
30× its usual pace: +60 this week vs ~2
+60 stars this week
Rank 40
Run the DeepSeek Harness AI agent natively on Android — no root, no Termux, extension center included | 手机本地运行 DeepSeek Agent，免 Root 免 Termux
15× its usual pace: +58 this week vs ~4
+58 stars this week
Rank 41
A curated, battle-tested AI-coding toolkit: Claude Code plugins, subagent orchestration, quality gates, and ready-to-copy prompts, extracted from real production use.
5× its usual pace: +482 this week vs ~94
+482 stars this week
Rank 42
Open-source orchestrator and ADE for running Claude Code, Codex, OpenCode, Pi and other AI coding agents in parallel — locally or 24/7 on your own server.
5× its usual pace: +98 this week vs ~20
+98 stars this week
Rank 44
ARGI [/ˈɑːr.dʒiː/]: Agent Runtime and Graph Intelligence 🚀
4× its usual pace: +35 this week vs ~9
+35 stars this week
Rank 45
A command-line tool for Z-Library. Search, download, send to Kindle, and integrate with AI agents.
3× its usual pace: +216 this week vs ~63
+216 stars this week
Rank 46
Open-source video pipeline (clipper, AI video & more) — scout trends, clip long videos, generate captioned shorts, design motion graphics. Local-first, BYOK. AGPL-3.0.
4× its usual pace: +34 this week vs ~8
+34 stars this week
Rank 47
Local observability for AI coding agents. Agenttrail Map shows project structure and activity; Agenttrail Kitchen visualizes tasks and role contributions in 3D.
3× its usual pace: +101 this week vs ~30
+101 stars this week
Rank 48
An auto-evolution framework that optimizes anything — your 7×24 team of algorithm engineers.
3× its usual pace: +630 this week vs ~192
+630 stars this week
Rank 49
通用、模块化的自媒体内容生产与经营 Skills / A modular, tool-agnostic self-media content skill suite
3× its usual pace: +36 this week vs ~11
+36 stars this week
Rank 50
The simplest, thinnest harness that gives an LLM complete freedom to control a Mac.
3× its usual pace: +52 this week vs ~16
+52 stars this week
Rank 51
ClearAI：以本体为纲的原生 DSH 插件 —— 经证据核验的研究长成可浏览的领域本体。 · ClearAI: an ontology-first DSH agent preset — evidence-checked research that grows into a domain ontology.
3× its usual pace: +452 this week vs ~140
+452 stars this week
Rank 52
Google Antigravity Standalone for Android - Autonomous ARM64 Client & Freeform Environment
6× its usual pace: +32 this week vs ~5
+32 stars this week
Rank 53
Evidence-grounded scientific agent system for research automation, citation verification and paper production.
+30 stars this week after quiet weeks
+30 stars this week
Rank 54
Grok Bot for Hyperliquid: an open-source 7-agent trading desk with roles, system prompts and 17 skills to research, size, execute and review. Also runs in Claude Code, Cursor, Codex and Gemini CLI.
3× its usual pace: +28 this week vs ~9
+28 stars this week
Rank 55
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.
2× its usual pace: +134 this week vs ~58
+134 stars this week
Rank 56
Bridge between QQ (SnowLuma OneBot v11) and DeepSeek Harness agents: social simulation, safe MCP tools, slang learning and more.
2× its usual pace: +282 this week vs ~126
+282 stars this week
Rank 57
Android Harness AI Coding Agent Native app for mobile, No Root No PC
2× its usual pace: +22 this week vs ~9
+22 stars this week
Rank 58
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.
2× its usual pace: +54 this week vs ~26
+54 stars this week