A Telegram bot for generating and managing temporary email inboxes on demand.
Official implementation of GeoNeXt: Video Generative Models as Geometry Learner (ECCV 2026).
ViRDM: Taming Representation Distribution Matching for Few-Step Causal Video Generation
Continual Learning Mechanisms Compose for Long-Horizon Memorization
A minimal coding agent harness that fits in the prompt
대한항공 공개 보너스 좌석 날짜 조회와 공식 달력 자동 입력을 지원하는 Windows·macOS 개인용 로컬 앱
PumpGuard AI: Telegram bot that screens new Robinhood Chain / hood.fun token launches with 4 LLM agents, a risk manager and a creator reputation book. Bring your own model: Grok by default, any OpenAI-compatible API works. Dry-run only.
一键诊断并修复 Codex / ChatGPT 桌面版「正在重新连接 1/5 → 5/5」| The Codex Rust backend ignores the Windows system proxy unless respect_system_proxy = true
整合型 Agent Skill 集合:一个技术生态或一类任务 = 一个 skill,581 条上游精编重写合成,仓库型来源固定 commit。
Unofficial study: Jev-style parallel typed decisions on stock 1.5B-8B models on an Apple Silicon laptop. Benchmarks, research notes, and a Hugging Face Space demo.
Qwen3.8-27B EXL3 (4.00 bpw) + DFlash2 speculative decoding for ExLlamaV3, validated at 262k context on a 24 GB RTX 3090
MCP server for TypeSafe Jev: typed classify, score, check, match and screen for any agent, with confidence on every answer
python sdk for dashscope [model studio](https://www.alibabacloud.com/en/product/modelstudio?_p_lc=1)
Metacognition for any agent. Drastically improves accuracy with almost no increase in cost.
—
Jev decision layer for Muse: a fast, cheap TypeSafe AI gate before expensive agent work — confidence policy, recipes, reference router, honest measurement.
Real-time Steam game leaderboards (最热游玩 · 热销 · 特惠 · 新品 · 免费游戏 · 限时免费) — live player counts, CNY prices & deals, auto-refresh every 60s.
微信 PC 端「按人分诊」悬浮面板:离线 OCR 读屏 + Jev 判断这一轮谁需要你出手。只判断不代聊、不注入微信、每次调用可回查。
—
Siemens TIA Portal PLC and HMI implementation for the DT_PSA_OPCV digital twin
Jev-like model: Qev, fine-tunes Qwen into a decision model
DevFlow AI:面向AI辅助开发的研发效能分析平台。原创协作链路、指标口径、样本提示与真实开发案例。
Local Yumesute preservation server based on server-of-dreams. Japanese and English setup guides. Early preview.
—
面向软件需求生命周期的质量评估与复盘 Agent — original rule engine, versioned evidence and human review
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
—
Argolink-focused SDK gateway compatibility and error mapping for DeepSeek V4 Pro.
Complete music production environment for Windows with piano roll, mixer, effects, synthesizers and pattern-based workflow for beatmakers and producers.
Local ComfyUI workflow for watermark, mosaic and irregular overlay removal with Qwen Image 2.1 and SAM
Fast and modern video editor for Windows with rich effects library, advanced timeline, auto-captions and smooth 4K export. Built for content creators and short-form video.
多租户 SaaS 智能客服:LangGraph 五技能诊断闭环 + 转人工真实流转(派单/接管/办结/回流)。FastAPI · PG 主从 · Redis · 本地向量化
⚖️ Kadı — a calibrated judge in front of your frontier LLM. Routine decisions answered locally with a conformal-prediction guarantee (≤ α disagreement with the oracle); everything else escalated and learned from. Zero-dependency Python.
Scholars Reading List search API: semantic and keyword search over ~200M papers, map positions, citation walk
🍔 麦门运势抽签点餐 Skill——先摇今日运势(大吉到平六级签等+签诗+宜忌),再随机搭配一整套麦当劳套餐;门店/价格/热量全部来自官方 MCP 实时数据。2026 麦当劳程序员创意开发大赛参赛作品 · 使用 WorkBuddy 开发
Dual-WM: DUAL-LATENT WORLD MODEL
IRC-style desktop client for MeshCore: chat on several nodes, bots and addons, firmware builder with OTA.
Procedural world maps from simplified plate tectonics: erosion, rivers, climate, biomes and trees. Python, NumPy, SciPy, Pygame viewer.
—
WearableQA A Benchmark for Health Reasoningover Real-World Wearable Data
Shared project context, work management and observable execution for humans and their coding agents.
Lightweight, async multi-agent framework with native SQLite persistence and handoffs for WhatsApp, Telegram, Webhooks, and CLI. Production-hardened with HMAC webhook security, automatic retries, and turn rollback. Supports OpenAI, Claude, Gemini, and local LLMs.
—
—
MobARK is a security research platform for Android and iOS apps. Use it for security review, pentesting support, or general AppSec work. The static analyzer handles APK and IPA files with androguard, Semgrep, and Gitleaks, plus an AI agent you can ask about the findings, using Ollama, LM Studio, or your own API key.
A lightweight API proxy for domestic and international CodeBuddy backends, providing OpenAI, Anthropic, and Responses-compatible APIs with native support for Claude Code, Codex CLI, and DeepSeek Harness.
A lightweight, production-ready RAG knowledge base system powered by LangChain, Ollama, and ChromaDB **PonyRAG** 是一个基于 RAG(检索增强生成)技术的本地知识库问答系统,专为企业和个人知识管理场景设计。系统完全本地部署,保护数据隐私,支持多种文档格式,提供智能问答和知识检索服务。
—
Evidence-governed autonomous experimentation and falsification for computational science
Forgex 是一款面向 Windows 平台的集成式渗透测试工具箱。它将信息收集、漏洞扫描、爆破利用、逆向分析、移动端 Hook、代理隧道等数十款主流安全工具聚合在一个统一的可视化界面中,单击即可启动,免去了手工配置环境、切换多套软件的繁琐
Dana | MCP Server with python and tailscale
Runs the complete released Drosophila male CNS connectome (165,122 neurons, 25.5M edges) inside a physical fly body, in closed loop, on one GPU.
[Preprint] TPPO: Tracking-Value PPO for Language Model Reasoning
Keep corrections and internal process away from ChatGPT/Codex output.
Aseprite automation for AI agents. Simple. Stable. Congruent. Create, edit, verify, and export pixel art and sprite animations via CLI, Agent Skill, or MCP server.
Kin 的小脑瓜:source-backed memory, affect, desires, personality and bounded exploration. Derived from MemoryPalace.
This repository provides the code and analysis pipelines used in this study, including the source code for the PIMFP2 web tool, which supports multiple input configurations, including unimodal and multimodal inputs as well as single-view and multi-view inputs, together with analysis pipelines for intramuscular fat (IMF) and fatty acid (FA) traits.
📚 An educational, step-by-step implementation of LLM KV cache reuse — learn prefix caching, chunk keys, storage & eviction, from pure Python to GPU tensors and vLLM integration. Inspired by LMCache.
设计Skill:面向 AI 生图与视觉设计的开源 Agent Skills | Design Skills: Open-source Agent Skills for AI image creation and visual design.
Query-based feed-forward 3DGS reconstruction.