Havuz

Local LLM inference on Xbox Series S in Developer Mode: llama.cpp (GGUF) + ONNX Runtime GenAI, an OpenAI-compatible LAN endpoint (off by def

Real-world percentile benchmark for LLM inference servers: 1% A/B resolution, offline HTML reports. I am now on X also: https://x.com/bkuype

Skills from StarSling (agent-native CI)

리버스 Nano Banana Ext API: 1K $0.0125. model ID `gemini-2.5-flash-image-preview`, OpenAI-compatible, async task + cost in every response, $1 m

Canonical, LLM-maintainable knowledge base of every Inner Circle Trader (ICT) concept 2016–2026. 226 concept files, machine-readable JSON bl

Distill a repeated Jev classification task into a local model, on the fly — same answers, your hardware.

DSH Android 客户端：内置 Termux 环境跑 Node，在 Node 上跑 DSH 引擎，WebView 加载官方 Web GUI

나노바나나 / 나노 바나나 — Nano Banana Pro prompts: tested image prompt recipes with example renders, aspect ratios, the model id (`gemini-3-pro-image

Grok Imagine Image 2.0 API (`grok-imagine-image-2.0`): model id, pricing, parameters and copy-paste examples for the OpenAI-compatible APIMa

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

SOC, incident response and SecOps for Linux servers in one agent: learns each host, watches it with zero-token sensors, triages with a small

리버스 Grok Imagine 1.5 Ext API: 1K $0.015. model ID `grok-imagine-1.5-apimart`, OpenAI-compatible, async task + cost in every response, $1 min

A safe local browser wizard for installing an OpenAI OAuth sidecar beside self-hosted n8n.

Independent Android client for Codex with on-device tools, live voice, and Claude/Gemini collaboration. No Termux or PC server required.

面向电商视觉生产的 Agent 技能库：26 个可组合的 skill.md，覆盖平铺图 → 模特上身 → 裂变套图 → 详情页 → 主图视频 → 投前质检 → 平台合规校验。不锁后端，dLazy / OpenAI / Gemini / fal / Replicate / 火山方舟

可视化 Agent 编排平台｜基于 DeepSeek Harness，支持 Workflow、Multi-Agent、元编排、MCP 与 Human-in-the-loop｜Visual AI Agent orchestration platform for workflows,

A headless TypeScript data table for React, Angular and Vue. 25 native UI-kit adapters, optional AI agents, pivot tables, editing, mobile ca

Create Instagram Reels, TikToks and YouTube Shorts with adaptive AI editing, captions and motion graphics. Connect FavStash for planning, pu

A ready-to-use Obsidian vault for running an AI agent (Claude, Codex, etc.) as a personal OS: skills, grounding notes, and a self-improving

Hybrid GPU + CPU inference for mixture-of-experts models, written in Rust down to the CUDA kernels. A drop-in llama-server.

Threads marketing skills for Claude Code and Codex: posts, multi-post threads, hooks, replies, and a weekly plan in your voice. Publish via

Local-first credential manager & validator for red teams / pentest：致力于 ALL IN ONE 的本地凭据管理与验证平台 keyrealm（凭域）。 允许将红队渗透测试中获取的散落在各处的 API Key、云厂商

Free local AI video clipper for Shorts, Reels and TikTok. Browser dashboard and CLI with resumable batch projects, editable captions/SRT, br

Chat with local and cloud AI

Spec-driven development for Claude Code, layered on OpenSpec. You read one spec; parallel agents ship it to a tested commit.

reverse-engineered FLUX Kontext Max API: 1K $0.064. model ID `flux-kontext-max`, OpenAI-compatible, async task + cost in every response, $1

Self-hosted, automated data broker removal tool (Incogni & DeleteMe alternative). Dispatches GDPR/CCPA requests from your own mailbox with a

Design-system and vocabulary drift, enforced. Tailwind themes become law, rogue values get receipts, words live in a database agents inherit

让 AI 自己编译、烧录、上板验证 STM32 固件，崩溃能定位到源码行 | Closed-loop STM32 development for AI coding agents

Veo 3.1 Fast API (`veo3.1-fast`): model id, pricing, parameters and copy-paste examples for the OpenAI-compatible APIMart endpoint.

NetEase Cloud Music Together for ChatGPT — an MCP/Bridge project with realtime room events, bidirectional chat, lyrics context, playback & q

FLUX 2 Max API (`flux-2-max`): model id, pricing, parameters and copy-paste examples for the OpenAI-compatible APIMart endpoint.

Reverse-Engineering Nano Banana 2 Ext API: 1K $0.015; 2K $0.02; 4K $0.025. model ID `gemini-3.1-flash-image-preview`, OpenAI-compatible, asy

reverse-engineered Grok Imagine 1.5 Edit Ext API: 1K $0.015. model ID `grok-imagine-1.5-edit-apimart`, OpenAI-compatible, async task + cost

Your Claude Code agents, living in a tiny pixel-art office. Desktop widget + browser app, with tokens, a shop and a lounge.

A Chromium fork built for AI agents: built-in MCP server, per-agent logged-in identities, sessions as tab groups

The video editor you never have to learn. An agent in Claude Code edits your videos on your Mac, and you leave notes from your phone.

本地优先的个人内容创作工作台：把资料、观点、核心稿、多平台版本、配图、排期与复盘串起来，一份稿子多平台生成爆款内容，热点监测，对标账号监控，数据分析，发布数据回流复盘，支持 X, 小红书，抖音，公众号，B 站，按需连接 AI 与平台服务。有打法有节奏的账号运营专家，目前不是完整版

A plugin that integrates PayPal's APIs and MCP server across multiple AI coding agents (Claude Code, Cursor and more) into your development

A background virtual screen for any Android AI agent: it operates apps on an invisible display while you keep using your phone. CLI / MCP /

Turn dry expert topics into high-engagement content with 7 core angles + 3 foundational lenses. An AI skill for any LLM, any niche, any lang

ChatGPT Subscription - a DeepSeek Harness plugin: bind your ChatGPT account via official OAuth and chat with ChatGPT models inside DSH, usin

Let Claude Code, Codex, OpenCode, Pi and other agents run DeepSeek Harness (dsh) sessions in parallel: research batches, material gathering,

开过的会，都听过。本地 AI 会议纪要：会议录音转文字、区分说话人、自动生成纪要和待办，支持腾讯会议/飞书/Zoom，录音不上传 · Local-first AI meeting notes for Mac: on-device transcription, speaker di

Offline emergency assistant: a local LLM and a 200 GB offline library on one Mac, with cited answers and no internet.

Cairn.ink Memory: lightweight cross-session memory for AI agents, with receipts.

🔒 Jail OpenCode, Claude Code, Copilot CLI and OmO in micro-VMs: only your project folders, only the networks you allow.

Free, open-source notch app for MacBooks: now playing, file shelf, calendar, Reminders, notes, shortcuts, timers, camera mirror, and your Cl

A visual editor for Mermaid diagrams — draw on a canvas, get Mermaid code. Local-first: no server, no accounts.

Packs text files into small images that an AI model reads for about half the input tokens. Includes a Claude Code plugin, a right-click tool