Havuz

Native Rapid-MLX provider for DeepSeek Harness (dsh) — dsh reads model facts from the server instead of your settings.yaml.

Desktop app for managing sandboxed OpenClaw and Hermes AI agent instances via Docker. Built with Tauri v2.

Wandesk — an open-source AI desktop. One shell, one AI kernel, one workerd; every app is a Worker site the built-in agent can write for you.

🦅 Multi-platform agentic AI harness — runs on Telegram, Discord, Teams & Terminal with a pluggable backend (Claude, Kilo, OpenCode, Codex, O

Add a talking 3D AI avatar to any web app. Samples and tools for the Perxona Connect SDK: <sv-presenter> web component, lip-sync, text-to-sp

Autonomous AI agent for Instagram — generates reels, carousels, memes, quote cards and posts them on autopilot. Free-tier LLM router, commer

Karpathy-style personal wiki for LLMs - markdown + frontmatter, no vector DB, no RAG. Built for Claude Code, works with any file-editing AI

🐚 The AI assistant that just works. One self-hosted app for every model you already pay for: Claude Code, Codex, Copilot, Gemini, Ollama and

Retrieval tools for coding agents, no model in the loop: SOTA-competitive files, search, read, code_pack and write_note over the folder your

Cognitive memory for LLM agents — local-first, and shareable across a team through the cloud.

餐饮连锁数字化 + AI 全景蓝图:原串(平价烤串连锁)的架构决策、业务口径、踩坑实录与可直接喂给 AI 的复刻指令。纯自然语言方案,不含代码与任何真实经营数据。

Open-source, voice-first AI mock interviewer. Upload your CV + a job description, practice out loud with an adaptive multilingual interviewe

LLM-managed personal knowledge base. Auto-extracts from git repos, Claude Code sessions (via ccrider FTS5), and iMessage bookmarks. Cross-pr

ИИ-агент (Claude, Cursor) читает данные рабочей базы 1С простым языком, без знания языка запросов — для бухгалтеров и аналитиков; и отдаёт с

Electron + Next.js desktop AI assistant that runs GGUF models locally using llama.cpp. Designed for offline use, portability, and zero-insta

Skill to generate the knowledge vault for projects using the Ralph loop

Claude Code / Codex / pi plugin that hands agent steps needing no text output to Jev (TypeSafe's judgment model) — measured p50 ~230 ms and

Optimize any AI agent’s skills, tools/MCP, and prompts against your own evals.

Local-first AI workbench for OCR, transcription, translation, image generation, mind maps and composable workflows. Built for DGX Spark.

👁️ The visual layer for CLI coding agents. Pipe JSON → interactive native UI → modified JSON back. Like fzf, but for visual decisions.

Turn a video — or an animated GIF, APNG, or WebP — into a timeline of still frames so an LLM can "watch" it as a series of images.

CodeMonk 🥋 —>An open-source AI powered code intelligence platform that helps developers understand, explore, and navigate complex codebases

把小说角色变成"活人"：每个角色是一个独立模拟体，只知道他该知道的，怕他该怕的，说他自己的话。 剧情线是引力场，不是铁轨——角色自由行动，世界只用"世界内的手段"把故事拉向既定节点。

Run several Claude Desktop accounts side by side on macOS, sharing Claude Code chats between them

Generative UI SDK for Swift (genui)

Zed extension for Headroom — context compression for AI agents

🐬 A pink qFlipper fork with LOTEI — a 100% local AI dolphin (Ollama) that chats, talks, watches your Flipper's screen, and one-click-install

Ready-to-run examples showing NoPII PII protection with OpenAI, Anthropic, LangChain, LlamaIndex, and more

Apple Icon Composer CLI & MCP server: create and manipulate .icon bundles and images with Liquid Glass rendering

Knowledge-graph RAG built from structured metadata instead of LLM extraction. 929M edges, zero LLM calls, 83.2% on held-out PubMedQA. Case s

Multi-Agent AI Desktop Client — 20 specialists auto-collaborate on your tasks. Visual Skill Store, PM orchestrator, streaming chat. Works wi

obra/superpowers extended — 14→24 skills, 6 GadaaLabs engineering improvements, override infrastructure, one-command install

[NeurIPS 2026] Official Implementation of "Visual-ERM: Reward Modeling for Visual Equivalence"

Memory as state, not search — deterministic, dependency-free memory for AI agents. The memory lives in a Markdown file.

Agent Harness for long-running agents that remember, continue, and learn from real work.

开源桌面 AI Agent 与 CLI：长期记忆、多 Agent、实时语音；Qwen3.8-27B Coder390 EfficientThink 本地 GGUF（llama.cpp Q2/Q3，内置 Q4 MTP）· Open-source desktop AI agent &

The most comprehensive MCP server for Meta's business platform. 200+ tools covering Facebook Pages, Instagram, Threads, Ads Manager, Commerc

"🐳 Dsh-Quant: The Everything-Plugin Ai native Quant OS "

The fastest inference engine for Qwen3.8-27B on the NVIDIA RTX 5090: up to 500 tokens/s for one agent and up to 2,000 tokens/s for many, con

Platform-agnostic AI agent: one Full-Stack Software Engineer brain, compiled into ready-to-paste adapters for ChatGPT, Claude, Copilot, Gemi

Русификатор игр в одну кнопку: Unity, Ren'Py, RPG Maker. Переводит файлы игры, а любую другую игру (Unreal, Godot) — поверх экрана. Офлайн-п

Agent 式的 SillyTavern 角色卡酒馆：高缓存命中，兼容酒馆角色卡与世界书，自带配图和一键朗读，角色形象固定，引导式创作角色卡。An agent-style SillyTavern card player for your own LLM API.

High-performance Knowledge Graph engine for AI, LLMs, and GraphRAG — built for the next generation of intelligent applications.

Keeps AI coding agents such as Claude Code, Codex, GitHub Copilot, Gemini CLI and Cursor to the least code that does the job and still works

Strands-based agents and harnesses for agentic benchmarks.

Daily AI brief for indie builders — what to ship, what to skip, who's eating whose lunch. Bilingual.

Tricorder provides simple, fast, focused feedback when hacking on Haskell (with or without LLM agents).

Use DeepSeek, GLM, Kimi, OpenRouter and more in Codex CLI and Claude Code: one local endpoint, API translation, failover. Single Rust binary

RHO: Evolving Agents in the Dark — Retrospective Harness Optimization via Self-Preference. Improving LLM agents from unlabeled past trajecto

Find and repair substance defects in AI-assisted prose, code, docs, and agent output. Reports defects, never authorship. Structural tests ov