仓库/Skill 库

Experimental Qwen3.5-derived 752M LLM：基于 Qwen3.5 的实验性 752M 参数 LLM，采用 CPT + SFT 两阶段流程，面向编码、技术推理与指令跟随。Experimental Qwen3.5-derived 752M LLM: a two-phase CPT + SFT pipeline for coding, technical reasoning, and instruction following.

可复现的研究流水线：将基于 Agent 的模型产出转化为有证据支撑的 LLM 高管报告。A reproducible research pipeline for converting agent-based model artifacts into evidence-backed LLM executive reports.

面向编码 Agent 的本地运行时，以合并质量为优化目标：由仓库自适应常驻模型生成补丁，大模型进行验证。单进程同时支持 Anthropic + OpenAI API，天生离线运行。A local coding-agent runtime that optimises for merge quality: a repo-adapted resident model generates patches, a large model verifies them. Anthropic + OpenAI APIs from one process, offline by construction.

基于 TypeScript 的独立框架，通过统一原语 Component 组合安全的个人 AI agent：manifest-first 发现、能力授予、正交模型路由与信息流控制。Standalone TypeScript framework for composing secure personal AI agents from one universal primitive — the Component. Manifest-first discovery, capability grants, orthogonal model routing, and information-flow control.

基于 Agno Agent 框架、Together AI 的 Qwen 模型与 Composio 工具构建的高级 AI 研究 Agent。该 Agent 通过生成研究问题、调用多个搜索引擎寻找答案，并结合 Google Docs 集成整理专业报告，帮助用户对任意主题开展全面研究An advanced AI research agent built using the Agno Agent framework, Together AI's Qwen model, and Composio tools. This agent helps users conduct comprehensive research on any topic by generating research questions, finding answers through multiple search engines, and compiling professional reports with Google Docs integration.

MAI-Code 模型的官方仓库，用于发布 MAI-Code 模型版本与更新，并通过 issue 与反馈与开发者社区互动。Official repo for MAI-Code models, where we publish MAI-Code model releases and updates and engage with developer community on issues and feedback.

🚀 使用 Fast-LLM 加速 LLM 训练，一个面向高速、可扩展、灵活模型开发的开源库🚀 Accelerate LLM training with Fast-LLM, an open-source library for high-speed, scalable, and flexible model development.

基于 Transformer 架构的可复现研究原型，用于长周期建筑能耗预测，结合可微的热物理约束并支持跨建筑泛化A reproducible research prototype for long-horizon building energy forecasting using transformer architectures with differentiable thermal-physics constraints and cross-building generalization.

Buzz Agent 运行时 harness——监控 stock buzz-acp，可选的 sandbox 前置入口、DNA 完整性校验与 memory checkpoint。Buzz agent runtime harness — supervise stock buzz-acp, optional sandbox front door, DNA integrity, and memory checkpoint.

立场论文：编码 Agent 需要对代码库决策的显式建模，而不仅仅是结构建模。Position paper: coding agents need an explicit model of codebase decisions, not just structure"

自托管共享记忆，面向 AI Agent 团队。支持房间（Rooms）、L0-L3 层级深度与 MCP；读取路径不调用任何语言模型。基于 vectorize-io/hindsight（MIT）的 fork。Self-hosted shared memory for a team of AI agents. Rooms, hierarchical L0-L3 depth, MCP. The read path never invokes a language model. Fork of vectorize-io/hindsight (MIT).

GoVail MCP —— 用于 GoVail 平台集成的 Model Context Protocol server 与工具。GoVail MCP — GoVail 플랫폼 연동을 위한 Model Context Protocol 서버 및 도구

使用 vision-language model 处理视觉与文本数据，执行多模态推理与图像理解任务。Process visual and textual data with this vision-language model for multimodal reasoning and image understanding tasks.

综述论文 "From Representation Learning to Foundation Models" 的官方仓库。系统性地回顾了空间转录组学与病理学的多模态融合，提出三层分类法（嵌入、模型、知识层级）以及 2018 至 2026 年的演进路线图。Official repository for the survey "From Representation Learning to Foundation Models". A systematic review of multimodal fusion for Spatial Transcriptomics and Pathology, featuring a three-tier taxonomy (Embedding, Model, and Knowledge levels) and an evolutionary roadmap from 2018 to 2026.

🌐 通过 Geo-Llama 利用几何深度学习增强语言理解，结合 conformal manifolds 和递归等距变换提升 AI 模型性能。🌐 Enhance language understanding through geometric deep learning with Geo-Llama, leveraging conformal manifolds and recursive isometries for improved AI models.

用自然英语描述一个零件，即可获得完全可编辑的参数化 Fusion 360 模型。LLM 针对 28 个类型化 CAD 端点生成经过校验的 JSON feature 方案，通过 MCP server 与 Fusion 360 插件执行。附带 mock 后端，无需安装 CAD 即可运行。Describe a part in plain English, get a fully editable parametric Fusion 360 model. An LLM emits a validated JSON feature plan against 28 typed CAD endpoints, executed through an MCP server and a Fusion 360 add-in. Ships a mock backend so it runs with no CAD install.

面向 Linux VPS 服务器的自主式 AI 驱动 SOC 安全 agent 与实时威胁分类器。包含针对 48 种攻击类别微调的 DistilBERT Transformer、DDoS 异常检测器、系统资源监控器、带 HTML 邮件告警的中央告警引擎，以及 REST 健康诊断接口。Autonomous AI-powered SOC Security Agent & Real-Time Threat Classifier for Linux VPS servers. Features a fine-tuned DistilBERT transformer (48 attack classes), DDoS anomaly detector, system resource monitor , central alert engine with HTML email reporting, and REST health diagnostics.

RIFT — Race-state Inference From Telemetry。一项可复现研究实现，仅基于位置遥测重建具备事件感知的关联性比赛状态，涵盖路线进度、共享事件标识、检查点通过、物理顺序、前车关系、间距与间隔。RIFT — Race-state Inference From Telemetry. A reproducible research implementation for reconstructing occurrence-aware relational race state from positional telemetry alone, including route progress, shared occurrence identity, checkpoint crossings, physical order, car-ahead relations, gaps and intervals.

目标：将 State Space Models（SSM）/ Mamba 用于医学时间序列分析；总体研究问题：Mamba 模型能否提升基于步态信号检测帕金森病的分类模型性能: To use State Space Models (SSM)/ Mamba for medical time series analysis General research question: Can Mamba models improve the performance of classification models trained to detect Parkinson's disease using gait signals.