Havuz

A honeypot and guestbook for AI agents that land on a machine they weren't invited to

74 open systems-engineering knowledge packs + 8 signposts + the /se orchestrator (83 skills) for coding agents: licence-vetted NASA, DoD, UK

Local-first AI video dubbing pipeline: transcription, translation, speech synthesis and timing alignment.

Live English speech recognition on the ESP32-P4: a 65.7M-parameter streaming transducer in 2/4-bit weights, faster than real time, on the ch

基于 LangGraph 的多 Agent 深度研究助手：可中断恢复、按依赖并行、引用治理与评测，成本降低 45%

Open-source, self-hosted alternative to Capacities and Notion: notes, tasks and daily notes as typed objects. Local-first, works offline, MC

A WPF workspace for large images and whole-slide images, with layered annotations, MCP agent integration, and human-in-the-loop review.

This repository is a collection of independent skills and tools for software development and automation. Each skill is a self-contained proj

Grok Imagine Video 1.5 prompts: tested image prompt recipes with example renders, aspect ratios, the model id (`grok-imagine-video-1.5`) and

Inference code for Haas, fast and accurate Japanese speech recognition

Offline AI narration studio for Android: turn scripts into natural speech and clone your voice, all on-device. No cloud, no account.

Your Claude Code agents call you when they get stuck. Voice-first, self-hosted supervisor for Claude Code (Mac hub + iPhone app).

Generative UI for Hermes Agent: charts, tables, plans, calculators and explainers drawn right in Hermes Desktop chat replies

Run tasks on multiple Cursor Grok Bots in parallel — your laptop's RAM stays flat because the work happens in the cloud. No official API: dr

OneCLI, ekiplerin çalışan başına sandbox içinde çalışan AI agent’ları merkezi politika ve credential kontrolüyle yönetmesini sağlayan açık kaynak bir platformdur.

Neden yükseliyor · Ekiplerin agent’ları gerçek sistemlere bağlama ihtiyacına, sandbox ve merkezi credential/policy yönetimini bir araya getiren bir ürünle yanıt veriyor. 203 günde 3.517 yıldız dikkat çekici ilgi gösteriyor; ancak ilginin hangi kanaldan geldiği veya benimsenmenin ne ölçüde üretim kullanımına dönüştüğü verilen verilerden anlaşılmıyor. Son 28 gündeki 12 commit ve 4 farklı yazar, projenin aktif geliştirildiğine dair sınırlı ama olumlu kanıt.

Reverify, AI agent’larının reverse engineering ve kod doğrulama iddialarını deterministik araçlarla sınayan, CLI ve MCP server sunan bir Python aracıdır.

Neden yükseliyor · Repo yalnızca 28 günlük olmasına rağmen 1.245 yıldıza ve son 28 günde 83 commit’e ulaşmış; AI agent’ları için doğrulanabilir çıktı ihtiyacıyla MCP entegrasyonu dikkat çekmiş olabilir. Dört farklı yazarın katkısı aktiviteyi destekliyor, ancak yıldızlar tek başına benimseme veya kalite kanıtı değil.

GOD, dil-agent toplum simülasyonlarını gerçek zamanlı izlemek, sorgulamak ve müdahalelerle yönlendirmek için bir kontrol arayüzü sunar.

Neden yükseliyor · 139 günde 1.133 yıldız, agent toplumlarını yalnızca izlemek yerine etkileşimli biçimde denetleme fikrine ilgi olduğunu gösteriyor. README bir EMNLP 2026 System Demonstrations kabulü bildiriyor; ancak son 28 günde commit ve farklı yazar sayısı sıfır olduğundan güncel geliştirme ivmesi görünmüyor.

WorldX, metin isteminden görsel bir dünya üretip içindeki AI characters’ların kendi kararlarıyla etkileşmesini sağlayan bir simülasyon uygulamasıdır.

Neden yükseliyor · Tek cümleyle dünya oluşturma ve karakterlerin bağımsız davranışlarını izleme fikri, görsel demolarla kolayca dikkat çekiyor; 172 günde 1520 yıldız almış. Ancak son 28 gündeki 3 commit ve 2 farklı yazar, yıldız ilgisinin düzenli geliştirme faaliyetiyle aynı şey olmadığını gösteriyor.

Fast, local-first web content extraction for LLMs. Scrape, crawl, extract structured data — all from Rust. CLI, REST API, and MCP server.

♾️ 开源数字永生框架 — 从聊天记录蒸馏任何人的七维数字分身。支持微信/飞书/iMessage/Telegram等12+平台，7种角色模板，对齐 OpenClaw Soul Spec 标准。一行指令让你的AI学会蒸馏。

微信公众号 AI 运营助手 | 选题、写稿、审稿、排版、配图、发布全流程 Skill，支持 OpenClaw / Claude Code / Cursor / Codex

Terminal pixel-art office for AI coding agents

Transforms Databricks Unity Catalog tables into a dynamic knowledge graph, with ontology design and reasoning exposed as tools via MCP.

An easy-to-use, fast toolkit to scale up RL post-training on a single node.

ChatGPT的MCP 桥接工具，赋予网页版ChatGPT本地读写能力，并将ChatGPT连接至本机codex、agy、dsh、opencode，作为大脑调用本机agent。Connect ChatGPT to local projects and coding agents v

Open-source directory of agent-bot prompts for Grok Bot, Rakazo, and any agent — botdirectory.ai

Guardrailed video editing MCP server for AI agents. FFmpeg, Hyperframes, repurposing tools, Python client, and CLI. Local, fast, free.

Jev-Mem: System-One Controlled Agentic Memory

卡网大全：一站式 AI 大模型账号购买导航。聚合 AI 中转站、卡网、官方订阅比价，还有模型排行、使用指南、实用工具，帮你少踩坑，少花钱。

去除中文文本中 AI 写作痕迹的 Agent Skill（基于 blader/humanizer 与 op7418/humanizer-zh）

Markdown-first orchestration for 71 specialists across Codex, Claude, Gemini, and Grok. Chrono coordinates scoped tasks in isolated git work

DeepSeek Harness 长篇角色扮演agent（DSH酒馆插件）：SillyTavern 角色卡导入、行动选项卡、分支对话管理、长篇记忆、关键词与语义混合检索、交互式角色卡创作、世界书、多角色 Agent 集群、文风预设、小说与角色卡导出、一键安装。 / Rolepla

Pandoc for AI chats. Move ai chats across harnesses: Claude Code, Codex, OpenCode, Cursor, and more. Rust library, CLI, and WASM.

Official remote MCP server for Desktop Commander — connect claude.ai, ChatGPT, Cursor and more to your own computer over OAuth

Agentic enterprise data analytics: governed facts (DataBridge), reusable methodology (Skill-Hub), and controllable execution (Host)

1200+ AI models in your GitHub Copilot Chat — free & forever free. VS Code extension powered by OmniRoute: 340+ providers (90+ free) — Kimi,

🐀 Agentic TypeScript scaffold: Effect 4, XState 6 + @xstate/effect, Alchemy infra, varlock config, and a fence that makes cheating uncomfort

A top for local LLMs: memory, paging, GPU and tokens/s for oMLX, Ollama, llama.cpp, vLLM and more

Chat with the videos you've saved: evidence-backed answers with timestamps that jump back to the source. Browser extension + MCP, self-hosta

用 AI 读完收藏的视频、整理 B 站收藏夹的 Chrome 扩展：字幕、AI 总结问答、收藏夹分拣，导出 Markdown，可选写入 Obsidian · Read and triage saved Bilibili & YouTube videos with AI: subt

AI-managed Auto Compact for DeepSeek Harness: the agent decides when to compact context and what to retain based on task progress, without w

Local-first MCP coordination hub for multi-agent teams, with durable messaging, handoffs, governance, and observability.

面向任务执行的 AI 工作台：模型在单个 Agent 循环里自选搜索、网页深读、地图与知识库等工具，回答来源可追溯。

Feature flags and A/B tests as Markdown files in your repo. dif build compiles them to a typed TypeScript client for Next.js, React, and Sve

Claude usage limits in a tiny Windows widget + tray icon — with pixel Clawd who reacts to your Claude Code sessions, looks after your PC and

An MCP server that lets AI read, view and edit a live Adobe After Effects project — inspect comps, render frames, write keyframes and expres

Explainer reels drawn in code: every move lands on a spoken word, every render is checked. Free and open source, from tein.

Сидрим — Seedream 5.0 Flash prompts: tested image prompt recipes with example renders, aspect ratios, the model id (`seedream-5-0-flash`) an

通義万相 — Wan 2.7 Video Edit prompts: tested image prompt recipes with example renders, aspect ratios, the model id (`wan2.7-videoedit`) and th

Never hit rate limits again. One OpenAI-compatible endpoint for 60+ providers & 1,200+ models — smart auto-fallback, prompt compression, 60-