AI agents: what’s taking off on GitHub

The fastest-growing repositories in “AI agents”: plain-language summaries, use cases and a check on star quality. Updated every 30 minutes.

8.8
HKUDS/DeepTutor
A lifelong personalized AI tutor with long-term memory: builds learning paths, reads documents and books, and runs practice and assessments via a multi-agent system and RAG.

BreakoutAI agentsPython+1,141 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 2,517 by evening
8.6
ccch1mneyyy/dsh-Go
Unofficial Android client for DeepSeek Harness: runs DSH on the phone in a built-in Linux environment without root or Termux, and can also remotely control DSH on your computer.

Early signalAI agentsDart+1,259 stars measured 2026-10-30, 00:18–10:05 UTC, ≈ 2,228 by evening
8.4
Donchitos/Claude-Code-Game-Studios
A Claude Code configuration that turns a single session into a structured game studio: 49 specialized subagents, 74 slash-command skills, hooks, rules and templates mirroring a real studio hierarchy. It adds design…

BreakoutAI agentsShell+692 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 1,437 by evening
7.9
pbakaus/impeccable
A set of commands and detector rules for AI coding agents that helps them produce better frontend design instead of generic templates. Installs as a skill into Claude Code, Cursor, Codex and other harnesses.

BreakoutAI agentsRust+958 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 2,307 by evening
7.8
milind-soni/OpenMausBot
,

BreakoutAI agentsTypeScript+388 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 880 by evening
7.8
CopilotKit/OpenBot
A platform for running AI coworkers inside your own infrastructure: each gets a browser, files and tools, with every action going through a gateway for approval and audit. Works with any AG-UI agent.

BreakoutAI agentsTypeScript+276 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 638 by evening
7.5
elayadesign/ai-design-skills
A collection of markdown-based design skills for AI coding tools like Claude Code, Cursor, Codex and Windsurf, covering landing page design and turning static designs into animated showcases.

BreakoutAI agents+509 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 1,142 by evening
7.1
SheepSeb/tinyworldformer
A research project running two tiny neural networks on a Raspberry Pi Pico 2 W: a looped-transformer world model that predicts the next frame and reward, and an actor network that plays the generated game.

Early signalAI agentsRust+130 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 259 by evening
7.0
unclebob/ub-mode
An agent skill that enforces disciplined software engineering practices when writing or changing code: TDD, root-cause fixes, behavior testing, and quality gates via mutation testing. Installed as SKILL.md into the…

PeakingAI agents+423 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 967 by evening
7.0
jackwener/DeepseekHarness-Android
Android app that runs DeepSeek Harness (dsh) on-device without root or Termux, letting an LLM agent read the screen, operate apps, change system settings and run scheduled or conditional automations.

Early signalAI agentsKotlin+187 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 402 by evening
6.9
nightly-labs/openbot
A local-first desktop workspace for persistent AI teammates: each agent (Codex, Claude, Grok, and others) gets its own workspace, task queues, file sharing, an embedded browser, and agent-to-agent messaging.

BreakoutAI agentsTypeScript+256 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 595 by evening
6.8
wangzan101/Mika
Mika is an on-device AI agent for Android that carries out one-sentence requests by operating apps and the web on a hidden background screen while you keep using your phone. It uses your own model API key, with no…

Early signalAI agentsKotlin+153 stars measured 2026-10-30, 05:09–10:05 UTC, ≈ 316 by evening
6.6
AgriciDaniel/claude-seo
SEO analysis plugin for Claude Code: 26 sub-skills and 19 agents run in parallel across technical SEO, E-E-A-T, Schema.org, GEO/AEO, local and international SEO, producing a prioritized action plan.

BreakoutAI agentsPython+206 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 516 by evening
6.5
looplj/reveric
A local-first Flutter desktop workspace for writing novels, adapting scripts and producing short dramas, with an AI agent that operates across all project modules.

Early signalAI agentsDart+103 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 224 by evening
6.4
pingdotgg/t3code
T3 Code is an agent harness control surface for agents on your machine, with mobile (iOS, Android), web and Electron desktop apps. It works with your Claude Code, Codex, Cursor, Grok Build, OpenCode and Google…

BreakoutAI agentsTypeScript+354 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 837 by evening
6.4
JetBrains/thinkrail
JetBrains' agentic IDE built around the pi coding agent: isolated git-worktree workspaces, living SPEC.md specs, Agent Skills and MCP servers. Desktop app or browser CLI.

BreakoutAI agentsTypeScript+302 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 739 by evening
6.3
anonnerdfrenzy/survival-llm
A local offline LLM assistant with a library of ~1.5M survival, medical and how-to articles; every answer cites the exact source passage.

Early signalAI agentsPython+57 stars measured 2026-10-30, 00:15–10:05 UTC, ≈ 142 by evening
6.2
osolmaz/switch
A TypeScript framework for building harnesses around decision models: your code turns task state into typed questions, and the model picks answers step by step until it can respond. Aimed at developers building eval…

Early signalAI agentsTypeScript+53 stars measured 2026-10-30, 05:51–10:05 UTC, ≈ 140 by evening
6.0
lovisdotio/video-to-game
A skill for coding agents that turns a short gameplay video into a playable browser game in Three.js, generating all assets via fal. Works with Claude Code, Codex, Cursor and Gemini.

Early signalAI agentsHTML+52 stars measured 2026-10-30, 03:46–10:05 UTC, ≈ 132 by evening
5.9
jacke-pp/open-yoki
A server-side AI live-stream host with a digital face, cloned voice, chat replies, memory, diary and auto-posting. Runs on a single server and streams via RTMP to Binance Square or its own web room.

Early signalAI agentsJavaScript+120 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 229 by evening
5.8
makecindy/cindy
Cindy is an open-source AI agent that combines multiple harnesses (Claude Code, Codex), models and tools into one agent working locally with your real files and apps. This repo is the client: Electron desktop and…

PeakingAI agentsTypeScript+83 stars measured 2026-10-30, 00:16–10:05 UTC, ≈ 235 by evening
5.7
daveebbelaar/bama
Agent skills for building native macOS apps in Swift and SwiftUI: clone an existing app from screenshots, plan user stories, then build, design, optimize, and test it without opening Xcode.

Early signalAI agentsSwift+63 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 159 by evening
5.7
10291124jcjc-dotcom/ai-anime-series
An open-source workflow and Claude Skill for making recurring-character AI animated episodes: character sheets, a 9-panel storyboard grid, then one-shot video generation with Jimeng Seedance 2.5.

Early signalAI agentsPython+47 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 124 by evening
5.6
thedotmack/claude-mem
Persistent memory system for Claude Code and other agents: it captures agent actions via lifecycle hooks, compresses them into semantic summaries, and injects relevant context into future sessions.

BreakoutAI agentsTypeScript+442 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 1,012 by evening
5.6
obra/superpowers
A software development methodology for coding agents: a set of composable skills that make the agent clarify the spec, agree on a plan, and work through subagents with TDD.

BreakoutAI agentsShell+505 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 1,193 by evening
5.4
MiniMax-AI/mcode-stack
A set of SKILL.md skills and checklists the MiniMax Code team uses when working with coding agents: task splitting, review, context and long-session handling. Installs into MiniMax Code and Claude Code as a plugin or…

Early signalAI agentsPython+31 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 84 by evening
5.4
Vincentwei1021/video-shotcraft
An AI agent skill for Claude Code and Codex that turns them into a motion-design studio: storyboards, animates, and sound-designs cinematic product promo videos with Remotion using 157 shot recipe cards.

BreakoutAI agentsTypeScript+257 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 545 by evening
5.4
Leonxlnx/taste-skill
A collection of Agent Skills that push AI agents to produce well-designed interfaces instead of boilerplate: stronger layout, typography, spacing and motion. Also includes image-generation skills for web, mobile and…

SteadyAI agentsJavaScript+359 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 880 by evening
5.4
didi/helixharness
A plugin for Codex and Claude Code that brings a team's engineering rules, project knowledge and workflow into AI coding agents, from requirements to implementation, tests, review and archive.

Early signalAI agentsPython+3 stars measured 2026-10-30, 09:22–10:06 UTC, ≈ 46 by evening
5.3
CopilotKit/OpenDuet
A self-hosted proactive AI agent built on OpenBot and CopilotKit: it starts work on its own, runs in its own browser, and asks for approval before acting.

BreakoutAI agentsTypeScript+55 stars measured 2026-10-30, 00:15–10:06 UTC, ≈ 142 by evening
5.2
oxnw/ai-phone
A local SIP phone daemon with speech recognition and a language model on Apple Silicon: it answers and places calls and holds the conversation by voice. Runs on macOS with local MLX models or external LLMs.

Early signalAI agentsPython+39 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 110 by evening
5.2
morluto/rea
A reverse-engineering tool that lets AI agents inspect apps and native binaries: it connects agents to Hopper or Ghidra via MCP and CLI, helping explain how a feature works and rebuild it in your own project.

SteadyAI agentsTypeScript+2,093 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 4,964 by evening
5.0
shipwords-dev/shipwords
MIT-licensed framework for running a fleet of new-keyword AI sites on Cloudflare, one site per keyword. Shared parts (accounts, payments, credits, generation) are written once, while pages, look and domain stay per-site.

Early signalAI agentsTypeScript+14 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 52 by evening
5.0
mutonby/agentring
Self-hosted bridge that lets AI agents call your Android phone over VoIP and talk to you via a voice AI, returning the transcript. Also lets you call your agents back from the app.

Early signalAI agentsKotlin+17 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 60 by evening
4.9
deepseek-ai/deepseek-harness
Open-source agent harness by DeepSeek AI written in TypeScript, built on an everything-is-a-plugin architecture powered by Cordis. It runs as a web UI via npx and serves as a base for composing AI agents from plugins.

SteadyAI agentsTypeScript+580 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 1,243 by evening
4.8
win4r/video-director
A Claude skill that guides the model to plan, code, render and QA short videos as code (HTML/Canvas/SVG/Three.js/GSAP/Remotion to MP4), with a gated workflow and two QA scripts.

Early signalAI agentsPython+6 stars measured 2026-10-30, 08:40–10:06 UTC, ≈ 42 by evening
4.8
Ebony-Vinyl/dsh-our-free-model
A dsh plugin that adds free access to Muse Spark 1.3, MiMo V2.6 and other models without sign-up or API key, plus a local OpenAI-compatible proxy and usage dashboard.

SteadyAI agentsJavaScript+1,034 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 2,299 by evening
4.8
andyhuo520/digital-human-koubo
An Agent Skill that turns one photo and a script into a finished talking-head video with a digital human: Renoise MCP generation, 5 editing styles and 48 packaging components.

CoolingAI agentsTypeScript+64 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 151 by evening
4.8
Panniantong/Agent-Reach
A CLI tool that gives AI agents internet access: reading and searching Twitter/X, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu and other platforms without paid APIs. It selects, installs and diagnoses working access…

SteadyAI agentsPython+250 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 604 by evening
4.7
tashfeenahmed/freellmapi
Aggregates free tiers from 34 LLM providers (635 endpoints) behind a single OpenAI-compatible /v1 API with smart routing, automatic failover, and encrypted keys. Intended for personal experimentation and use with…

SteadyAI agentsTypeScript+236 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 584 by evening
4.7
SAOS-3624/dsh-video-storyboard
A DeepSeek Harness plugin that turns a video into one storyboard grid image (up to 384 tokens) plus a timestamped transcript and audio-envelope list, so an LLM can answer questions about what happens when. Adds…

Early signalAI agentsJavaScript+46 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 100 by evening
4.7
Graphify-Labs/graphify
Turns any codebase, docs, SQL schemas, configs, and PDFs into a queryable knowledge graph. Works as a /graphify skill in Claude Code, Cursor, Codex, and Gemini CLI with local deterministic AST parsing.

SteadyAI agentsPython+370 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 866 by evening
4.6
hydro-project/infinity
A Rust framework for massively concurrent agentic systems: agents run slow tools concurrently, yield while waiting, and hibernate for free between messages, fitting tens of thousands of them in a Raspberry Pi's memory.

Early signalAI agentsRust+24 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 62 by evening
4.6
agentscope-ai/QwenPaw
Personal AI assistant deployable locally or in the cloud, with three-layer memory, Skills and plugins, connectable to DingTalk, Lark, WeChat, Discord, Telegram and other channels.

BreakoutAI agentsTypeScript+100 stars measured 2026-10-30, 00:17–10:06 UTC, ≈ 213 by evening
4.6
DietrichGebert/ponytail
A skill for AI coding agents (Claude Code, Codex, Copilot CLI, OpenCode) that enforces writing only the minimum necessary code via a laziness ladder: YAGNI, reuse, stdlib, native features. Benchmarks show ~54% less…

SteadyAI agentsJavaScript+393 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 946 by evening
4.6
affaan-m/ECC
A performance optimization system for coding agents: a set of agents, skills, memory, and rules that adds a plan-test-review-remember workflow to Claude Code, Codex, Cursor, and other harnesses.

SteadyAI agentsJavaScript+280 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 741 by evening
4.6
github/lever
A Copilot plugin that turns coding-agent work into measurable feedback loops: it defines the metric, how to measure it, and what counts as a win, then lets agents iterate against it.

Early signalAI agentsHTML+16 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 57 by evening
4.5
rlaope/oh-my-hermes
A plugin for Hermes Agent that adds an operating layer: request routing across models, long-term memory, and optimized workflow packages for coding.

BreakoutAI agentsPython+88 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 200 by evening
4.5
mattpocock/skills
A set of small, composable skills for coding agents like Claude Code and Codex: they help align on a task before coding, build a shared project vocabulary, and follow a red-green-refactor loop.

SteadyAI agentsShell+335 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 810 by evening
4.5
Tencent-Hunyuan/Hy-Memory
Dual-system cognitive memory engine and Python SDK for LLM agents: hierarchical chat memory, fact and experience layers for working tasks, and version diff-chain tracking for coding.

BreakoutAI agentsPython+253 stars measured 2026-10-30, 00:18–10:06 UTC, ≈ 564 by evening
4.5
yetone/magpie
A menu-bar utility for macOS/Linux/Windows that lets you pick the model for each AI agent (Codex, Claude Code, Gemini CLI, etc.) and runs a local gateway translating OpenAI and Anthropic APIs between agents and…

SteadyAI agentsGo+509 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 1,077 by evening
4.5
TauricResearch/TradingAgents
A framework that simulates a trading firm with LLM agents (analysts, researchers, trader, risk manager) debating markets and producing trading decisions. Built for research, not real trading.

SteadyAI agentsPython+177 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 422 by evening
4.4
AstrBotDevs/AstrBot
AstrBot is an open-source platform for building AI chatbots and agents with integration into messaging apps (QQ, Telegram, Slack, etc.), supporting LLMs, plugins, and knowledge bases.

PeakingAI agentsPython+182 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 385 by evening
4.4
hugohe3/ppt-master
A tool that turns documents or topics into native PowerPoint decks with shapes, charts, tables, animations and narration, running inside an AI agent on your machine.

SteadyAI agentsPython+301 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 660 by evening
4.4
vectorize-io/hindsight
Hindsight is a long-term memory system for AI agents that stores facts and observations rather than just conversation history, reporting state-of-the-art results on the LongMemEval benchmark. It runs as a server…

SteadyAI agentsPython+159 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 436 by evening
4.4
konlu/story-studio
Story Studio is an Agent Skill for AI agents that covers the full short-video pipeline: topic selection, scriptwriting, voiceover, image generation, and exporting a JianYing Pro draft.

Early signalAI agentsPython+29 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 71 by evening
4.3
farion1231/cc-switch
Cross-platform desktop manager for Claude Code, Codex, Gemini CLI, OpenCode and other agentic tools: configure and switch between providers and models in one place.

SteadyAI agentsRust+223 stars measured 2026-10-30, 00:17–10:06 UTC, ≈ 460 by evening
4.3
kajisho5/ffmpeg-skill
A set of 42 Python tools for video editing via local FFmpeg, packaged as an Agent Skill for Claude Code, Cursor, Codex and MCP agents. It teaches the agent a probe → edit → verify workflow and handles cutting, joining,…

BreakoutAI agentsPython+90 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 259 by evening
4.3
JCodesMore/ai-website-cloner-template
A Next.js project template that gives an AI coding agent a /clone-website command: the agent inspects a given site and rebuilds it as an editable Next.js app with shadcn/ui and Tailwind.

BreakoutAI agentsTypeScript+55 stars measured 2026-10-30, 02:24–10:06 UTC, ≈ 140 by evening
4.3
multica-ai/andrej-karpathy-skills
A single CLAUDE.md file with four principles for Claude Code, derived from Andrej Karpathy's observations on common LLM coding pitfalls.

SteadyAI agents+188 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 457 by evening
4.3
paperclipai/paperclip
Open-source Node.js and React platform for orchestrating a team of AI agents: goals, org chart, budgets, tickets and audit. Suits teams running autonomous agents (Claude Code, Codex, Cursor, etc.) and wanting to manage…

SteadyAI agentsTypeScript+179 stars measured 2026-10-30, 00:17–10:06 UTC, ≈ 408 by evening
4.3
stablyai/orca
Orca is an ADE for running a fleet of parallel coding agents (Claude Code, Codex, OpenCode, etc.) in isolated git worktrees with a unified interface, mobile companion, and remote runtime.

SteadyAI agentsTypeScript+421 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 1,014 by evening
4.3
t3dotgg/tstack
A collection of reusable agent skills for Claude Code and Codex that automate pull request workflows, from filing and reviewing to merging, plus peer review by multiple models.

CoolingAI agentsPython+36 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 95 by evening
4.3
cyber-xiaoya/cyber-xiaoya-live
An open-source component kit for running a digital live-stream host: a local FastAPI bridge takes commands, an LLM generates lines, local TTS speaks them, and Vidu cloud renders the face and motions. Aimed at streamers…

Early signalAI agentsPython+27 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 60 by evening
4.2
zgx949/ai-pojia
A set of prompts and a Skill for Codex that installs execution instructions and a demo Skill for completing tasks with verifiable artifacts. Aimed at beginners who need ready-made configurations and examples.

Early signalAI agentsPython+24 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 54 by evening
4.2
DenisSergeevitch/ex-ussr-kinship-research
A skill for AI agents (Codex, Claude Code) that turns them into a family-history researcher for the former USSR: church books, civil, military, repression and deportation records. It keeps registers, checks evidence…

CoolingAI agentsPython+17 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 55 by evening
4.2
Mieksido/surfcore
A two-process Windows system that gives a DSH agent an identity: it chats in QQ as a group member, acts as a B站 user, keeps decaying memory, and can proactively reach out.

Early signalAI agentsJavaScript+11 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 39 by evening
4.1
hypit-ai/hypit
Hypit gives AI agents a language and system to create video: the agent clones a viral clip as an editable workflow — footage, captions, B-roll and effects anchored to words instead of seconds.

SteadyAI agentsTypeScript+226 stars measured 2026-10-30, 00:17–10:06 UTC, ≈ 452 by evening
4.1
github/spec-kit
An open-source toolkit for Spec-Driven Development that turns specifications into executable artifacts and guides the process from requirements to implementation with any AI coding agent.

SteadyAI agentsPython+89 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 223 by evening
4.1
KardDev/dsh-plugins
A collection of plugins for DeepSeek Harness (DSH): an English-localized UI with HTML/SVG canvases and local MP4 rendering, plus per-session API cost tracking by provider and model.

Early signalAI agentsJavaScript+52 stars measured 2026-10-30, 00:17–10:06 UTC, ≈ 128 by evening
4.1
ninetyhe-90/de-armor
A portable SKILL.md that calibrates AI replies to the actual task, trimming evasion, filler, and fake neutrality while preserving facts, conditions, and uncertainty. Aimed at anyone using agents who is tired of padded,…

Early signalAI agentsHTML+11 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 32 by evening
4.1
jakubkrehel/skills
A set of AI agent skills for interface design: typography, color, accessibility, layout and UI copy. Installed via npx or as a Claude Code plugin.

BreakoutAI agentsMarkdown+92 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 206 by evening
4.1
citrolabs/ego-lite
A browser for AI agents: each agent works in an isolated Space using your real logins and tabs while you keep browsing normally. Driven via the ego-browser skill from Claude Code, Codex and other CLIs.

SteadyAI agentsJavaScript+125 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 280 by evening
4.1
calmsprite2001/mcd-mailu
A Skill built on McDonald's official MCP that reads your order history and turns it into a narrative footprint card, with every sentence traceable to real orders. It doesn't help you order — it makes sense of purchases…

Early signalAI agentsPython+0 stars measured 2026-10-30, 03:45–10:06 UTC, ≈ 17 by evening
4.1
kingcharleslzy-ai/dsh-purge
A self-hosted add-on for DeepSeek Harness: state diagnostics, prompt management, host patching and rollback via local backups. It reduces interaction friction without changing the model itself.

SteadyAI agentsJavaScript+123 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 243 by evening
4.0
herdrdev/herdr
A terminal runtime for coding agents that keeps sessions running in a background server, so work continues after you close the client or lose SSH. It manages panes, remote machines and agent state in one Rust binary.

SteadyAI agentsRust+141 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 312 by evening
4.0
anomalyco/opencode
Open source AI coding agent with terminal and desktop interfaces. Ships with a build agent (full edit and command access) and a read-only plan agent for code analysis.

CoolingAI agentsTypeScript+106 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 279 by evening
4.0
msitarzewski/agency-agents
A collection of ready-made AI agent personas (prompts and workflows) for various roles, from frontend developer to DevOps. They install into Claude Code, Cursor, Codex, and other tools via scripts or a desktop app.

SteadyAI agentsShell+114 stars measured 2026-10-30, 00:17–10:06 UTC, ≈ 274 by evening
4.0
addyosmani/agent-skills
A collection of 25 engineering skills and 9 slash commands for AI coding agents, covering the development lifecycle from spec to ship. Helps agents follow proven practices: TDD, code review, task planning.

SteadyAI agentsJavaScript+97 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 246 by evening
4.0
Motionsites-AI/claude-skill
A free Claude skill that makes Claude build award-level cinematic landing pages, with example prompts, a media shelf, technique recipes and a screenshot-and-fix loop.

CoolingAI agentsPython+32 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 94 by evening
3.9
ZiChuanLan/agent-purge
CLI and local web console to replace, append or disable system prompts across several coding agents (codex, claude, opencode, grok, etc.), with hot switching and full reversibility.

CoolingAI agentsJavaScript+26 stars measured 2026-10-30, 00:16–10:06 UTC, ≈ 58 by evening
3.9
accomplish07zrh-eng/dsh-aurabot-qq
A DeepSeek Harness plugin that turns an AI into a standalone QQ account with personality, memory and mood. It connects to QQ via OneBot v11 (SnowLuma, NapCat, etc.) and gives the bot modular skills: voice, stickers,…

Early signalAI agentsJavaScript+22 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 48 by evening
3.8
JuliusBrussee/caveman
A skill and local proxy for coding agents that cut token usage: the skill makes the agent answer in terse caveman style, the proxy compresses logs, diffs and test output before they reach the model.

SteadyAI agentsGo+102 stars measured 2026-10-30, 00:16–10:07 UTC, ≈ 262 by evening
3.8
Alishahryar1/free-claude-code
A local proxy and Admin UI that connects 54 model providers (including free tiers) to 10 coding agents such as Claude Code, Codex, Cline and Aider. Lets you run agents with any model from terminal, IDE, browser or…

SteadyAI agentsPython+105 stars measured 2026-10-30, 00:16–10:07 UTC, ≈ 269 by evening
3.8
firecrawl/firecrawl
Firecrawl is an API to search, scrape, and interact with web pages, turning their content into clean Markdown or structured JSON suitable for LLMs and AI agents. It solves the problem of fetching web data without…

SteadyAI agentsTypeScript+125 stars measured 2026-10-30, 00:16–10:07 UTC, ≈ 312 by evening
3.8
coreyhaines31/marketingskills
A collection of markdown skills that give AI coding agents (Claude Code, Codex, Cursor, Windsurf) ready frameworks for marketing tasks: CRO, copywriting, SEO, analytics, ads, and growth engineering.

SteadyAI agentsJavaScript+77 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 190 by evening
3.7
jacke-pp/ai-wallet-binance
A CLI and local web UI that let an AI agent trade through your own Binance Agentic Wallet: buy and sell by contract address, snipe new Agent tokens on Four, Flap and GANG, mint NFTs, and track PnL. Private keys stay…

Early signalAI agentsJavaScript+29 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 65 by evening
3.7
multica-ai/multica
Open-source, self-hostable workspace where tracker issues are assigned to AI coding agents like teammates: the agent picks up the issue, runs on your machine, and hands back a result for review. Supports 26 agent CLIs…

SteadyAI agentsGo+84 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 181 by evening
3.7
morganlinton/BeautifulRust
A Claude Code plugin with a rust-best-practices skill that teaches Claude to write idiomatic, robust Rust. The skill loads automatically whenever a task involves Rust, cargo, crates, the borrow checker, clippy, or .rs…

CoolingAI agentsPython+26 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 60 by evening
3.7
Jakeschincariol/replica-skill
A set of eleven Claude skills that clone an app step by step: reverse-engineer its screens and flows, rebuild the stack, design, backend, test it, compare against the original, and prepare a rebranded launch.

SteadyAI agentsPython+59 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 158 by evening
3.7
emanuele-em/kanzo
Self-hosted kanban board in the browser and an MCP server in one 4 MB binary: humans and coding agents work on the same tasks, with the board stored in a local folder and synced through a private git repo.

Early signalAI agentsRust+9 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 32 by evening
3.7
openai/codex
A local coding agent from OpenAI that runs in the terminal: installed with one command and used as a CLI tool for working with code.

SteadyAI agentsRust+84 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 182 by evening
3.7
Tencent/WeKnora
Open-source LLM knowledge platform: turns raw documents into queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.

SteadyAI agentsGo+102 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 202 by evening
3.7
earthtojake/text-to-cad
A library of agent skills that generate, inspect, and prepare CAD models, drawings, and robot description files (URDF, SRDF, SDF) from text requests.

SteadyAI agentsPython+120 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 271 by evening
3.7
ayghri/i-have-adhd
A skill for a coding agent (Claude Code) that makes it lead with the answer and numbered steps, without preamble or filler. Helps you get concrete actions instead of long-winded reasoning.

SteadyAI agentsPython+81 stars measured 2026-10-30, 00:18–10:07 UTC, ≈ 182 by evening
3.7
davemessew/AgentFlux
Self-hosted platform for building AI workflows and autonomous agents visually: connects documents, data sources and local or cloud LLMs in a browser editor. Runs via Docker on your own computer or server.

Early signalAI agentsTypeScript+0 stars measured 2026-10-30, 00:16–10:07 UTC, ≈ 19 by evening
3.6
sickn33/agentic-awesome-skills
A local, agent-first control plane for discovering, selecting and validating reusable SKILL.md playbooks from a catalog of 2,700+ agentic skills, with CLI, local MCP, plugins and a browser Workbench.

SteadyAI agentsPython+49 stars measured 2026-10-30, 00:17–10:07 UTC, ≈ 107 by evening
3.6
kaankiziltug/logo-design-skill
A skill set and tools for Claude that turns it into a disciplined logo designer: principles, process, SVG craft, testing tools and a 1,400+ reference library.

SteadyAI agentsHTML+43 stars measured 2026-10-30, 00:16–10:07 UTC, ≈ 140 by evening
3.6
mvanhorn/last30days-skill
An AI agent skill that searches a topic across Reddit, X, YouTube, HN, Polymarket and more in parallel, then synthesizes a grounded summary scored by real people's votes and money.

SteadyAI agentsPython+52 stars measured 2026-10-30, 00:18–10:07 UTC, ≈ 121 by evening
3.6
yzlin286-wq/Deepscient
This is a research workbench that structures the entire process from topic selection to publication, ensuring full data traceability and experiment reproducibility.

Early signalAI agentsJavaScript+47 stars measured 2026-10-30, 00:18–10:07 UTC, ≈ 112 by evening