Havuz

DeepSeek Harness (dsh) 插件：注入系统提示词，强制 Agent 的思维链 / 规划 / 工具推理全程使用简体中文，防止中文思考漂移为英文。零外部依赖，任意本地目录 link 即装。

Real-time Windows system-audio translation captions for Chinese, Japanese, English, and Korean, with optional LLM revision.

Ground truth for enterprise AI: your AI coding tools, your accounts and every session on your own server, from one web console, CLI or API.

Open-source desktop coding agent for macOS. Use any model: local with Ollama, your own API keys or a ChatGPT sign-in. Parallel chats, MCP co

DeepSeek Harness (DSH) —— DeepSeek AI 开源的 Agent 智能体框架，群晖 Synology DSM / 飞牛 fnOS NAS 原生打包（SPK/FPK），无需 Docker，开箱即用。Web UI 管理、多模型、内置反向代理与局域网安全限

Personalized top-conference recommendations, human-guided paper notes, and shareable research galleries that grow with your work.

A modular RAG pipeline library for .NET, built on Microsoft.Extensions.AI abstractions

🤖 Bony Agent — Full-stack AI content production & distribution platform. One person can run an entire team: AI chat · image/video/music gene

A guide that teaches an AI coding assistant DeepSeek's superagent pattern and how to set it up with Claude Code or OpenCode

Desktop VRM companion that gives any OpenAI-compatible agent backend a body, voice, and presence.

Zusia, a study buddy inside Zotero: ask Claude Code, Codex or Antigravity about the paper you are reading. LaTeX, drawings, theorem links, g

VRChat向けのAIエージェント実装です。音声認識、ローカルLLM、VOICEVOX系TTS、VRChat OSC、YOLO/Pose、VLM、長期記憶、感情/欲求モデル、自律行動を組み合わせ、会話だけでなくアバターの移動、視線、表情、エモート、Use/Grab、ピアノ演奏まで

a truly open experimental coding harness

Self-hosted DNS filtering resolver with classic blocklist ad-blocking, AI-assisted domain classification via a remote LLM, and a web managem

Daily tracker of AI repositories trending on GitHub, with weekly analysis of rising topics and dominant languages, charts and CSV data, auto

Towards AI-Generated and Formally Verified Database Systems

Av. Baran Doğan'ın makaleleri — yazarın izniyle, atıflı ve ticari olmayan amaçla, hukuk alanlarına göre sınıflandırılmış yapay zekâ dostu ar

大模型故障 token（glitch token）资料库：收录各 LLM 分词器中已公开记录的异常 token，附异常表现、来源与复现链接。A curated catalog of LLM glitch tokens.

Compact coding-agent context without summarizing it away. Verbatim, offline, with a receipt and recall for every cut. CLI, MCP server, Claud

dsh-model-master（模型选择大师）— model selector for DeepSeek Harness: search across providers, probe any model with a real streaming request to mea

Lossless context folding for the Pi coding agent: only the oldest context changes; the fresh window stays untouched.

An experimental JEV-powered framework for forecasting short-term stock price direction from structured market data.

Local-first desktop app for AI agent teams. Runs with Ollama or cloud models. Mac + Windows.

A self-hosted Cloud API aggregator and control plane for managing LLM inference infrastructure across multiple machines.

Local-first creative intelligence for mobile-game UA teams — decompose creatives into structure, link them into a graph, turn every human ru

Edit a DeepSeek Harness (dsh) custom mode: system prompt, base mode, and plugin switches.

Hermes plugin: one /crew message becomes verified work on the kanban board. A coordinator owns every card until its proof passes.

Редактура українського тексту: прибирає сліди ШІ, канцелярит, суржик і кальки, зберігаючи зміст і голос автора. 63 маркери, 30 заборон, скан

Session Recorder by Proofline. Free, open-source Chrome bug reports with annotations, voice, screenshots and one-click GitHub issues.

One prompt → a swarm of 150+ specialist Cursor agents that plan, build, test & harden your whole project in parallel, sharing an Obsidian br

dsh-dock · DeepSeek Harness 功能坞插件：一张面板统一注册/开关所有小功能——用量记账（自定义单价·分时价）、模型设置与余额、19 种任务动画、任务通知（钉钉/飞书）、运行状态、图片理解代理、账号密码远程访问、10 款小游戏。每个功能独立模块，可开关、错

Multi-model routing for any AI coding agent: send each task to the cheapest model that can do it well. Works with Claude Code, Codex, Cursor

Point at any part of your running app, describe the change, and Lasso edits the real source file — not the DOM. Works with Vite and Next.js.

Local-first work memory for Claude Code, Codex and Copilot. Plain Markdown files hold your projects, people, decisions and meetings, with a

MCP server for Cisco Packet Tracer 9 — let Claude/ChatGPT design, build and validate Cisco network topologies in real time (OSPF, BGP, VLAN,

Agent Orchestrator App

Streaming privacy for LLM apps: pii-leak-benchmark (does your gateway leak PII in streams?), chunk-invariance (one-line test for split-bound

Agente de programación para la terminal, gratis y en español: usa las IA gratuitas de más de 20 proveedores (Gemini, NVIDIA, Cerebras, OpenR

🎯 Free & open-source AI interview copilot. Real-time answers from voice + screenshots for coding and job interviews. Works with Groq, Claude

A local agent workstation: one Go loop, three doors: Wails desktop, browser, and CLI. It writes code, reviews hunks, opens an isolated brows

Open-source AI dating agent — bring your own Tinder token and any LLM (OpenAI, Claude, Gemini, DeepSeek, Qwen, and more). Screens dates, mak

Astrology MCP server & Agent Skill — birth charts, transits, synastry across Western, Vedic/Jyotish & Chinese BaZi. Deterministic zero-depen

Local-first AI interactive fiction app — a self-hosted narrative engine with an async multi-agent brain that keeps story state in plain file

Production-ready cross-border e-commerce multi-agent system built on AgentScope 2.0 with DDD onion architecture, GatewayThrottle, semantic c

Automatic movie clipping for YouTube Shorts. Scene detection, LLM scoring, face tracking, subtitles, vertical crop.

DeepSeek Harness 插件：让自定义 llm-pi-ai provider 伪装成 Claude Code / Codex 客户端（伪造客户端身份请求头）。A DeepSeek Harness plugin: masquerade a custom llm-pi-ai

VoiceMem, voice agent’lara ses ve metinden bilgi ile duygu odaklı uzun süreli bellek sağlayan bir Python kütüphanesidir.

Neden yükseliyor · 41 günde 2.111 yıldız alması görünür ilgiye işaret ediyor; son 28 gündeki 49 commit ve 5 farklı yazar da geliştirme faaliyetinin sürdüğünü gösteriyor. Ancak yıldızlar gerçek kullanımın veya performans iddialarının doğrulaması değil.

mu, coding agent kararlarının bir kısmını küçük bir judge modeline devrederek pi tabanlı CLI ve desktop agent sunar.

Neden yükseliyor · 14 günlük yaşına rağmen 386 yıldız, son 28 günde 731 commit ve 11 farklı yazar raporlanıyor; bu kısa sürede dikkat ve geliştirme hareketliliğine işaret ediyor. Ancak bu veriler kalıcı benimseme veya üretim olgunluğunu tek başına kanıtlamıyor.

MAC, doğal dildeki tasarım brief’lerinden düzenlenebilir CAD parçaları ve deneysel olarak çok parçalı assembly’ler üretir.

Neden yükseliyor · 59 günde 1006 yıldıza ulaşması, doğal dilden CAD üretimine yönelik ilgiyi ve görsel olarak etkileyici parça/assembly demolarını yansıtıyor. Ancak README, assembly örneklerinin seçilmiş çıktılar olduğunu ve ölçülmüş prompt başarı oranı sunmadığını belirtiyor; 28 günde 22 commit olsa da farklı yazar sayısının 1 olması benimseme ve sürdürülebilirlik kanıtını sınırlıyor.

Tura, coding agent’ların birden çok araç adımını tek bir runtime-managed command graph içinde çalıştırarak tekrar eden LLM çağrılarını azaltmayı hedefleyen açık kaynak bir agent runtime harness’tir.

Neden yükseliyor · 83 günde 644 yıldız ve son 28 günde 18 commit dikkat çekiyor; token tasarrufu vaadini karşılaştırmalı benchmark’larla desteklemesi de ilgiyi artırmış olabilir. Ancak son 28 gündeki 2 farklı yazar, katkı tabanının hâlâ sınırlı olduğuna işaret ediyor.