仓库/Skill 库

2026 掌握 LLM 搜索：完整的语义 AI 手册Master LLM Search in 2026: The Complete Semantic AI Handbook

与他人协作构建——双方各自使用自己的 Claude、运行在各自的机器上。两个 Agent，一份计划，任务跨机器交接。Build something with another person — each of you with your own Claude, on your own machine. Two agents, one plan, tasks handed across machines.

基于 MCP 的实时来源声明与引用验证，支持 FIGI 解析与签名投递回执。Live-source claim and citation verification over MCP, with FIGI resolution and signed delivery receipts.

2026 AI 驱动的加密货币 Alpha 引擎 🚀——追踪叙事与空投机会。AI-Powered Crypto Alpha Engine 2026 🚀 - Sniping Narratives & Airdrops

面向 LLM Agent 的确定性写入时记忆整合与预取召回：采用本地模型，召回路径与整合决策中均不含 LLM。Deterministic write-time memory consolidation and pre-fetch recall for LLM agents. Local models, no LLM in the recall path or the consolidation decision.

使用纯 C99 MoE 推理引擎在 CPU 上原生运行 DeepSeek-V4-Flash-0731，无需 GPU、CUDA 或 PyTorch。Run native DeepSeek-V4-Flash-0731 on CPU with a pure C99 MoE inference engine — no GPU, CUDA, or PyTorch needed.

面向生物医学系统综述的 Claude Code skills 协议：MCP 锚定引用、实证完整性、PRISMA/PROSPERO 系统综述、新颖性核查、Zotero 操作、并行审稿手稿修订，并提供面向健康科学研究策划的生信、临床与科学传播 skill 集合。Claude Code skills protocol for biomedical systematic review: MCP-grounded citations, empirical integrity, PRISMA/PROSPERO systematic reviews, novelty checks, Zotero operations, and parallel-critic manuscript revision — plus a curated set of bioinformatics, clinical, and scientific-communication skills for health-science research.

Stacklit：AI Codebase Indexing 2026 — 为 Agent 提供零配置本地仓库映射Stacklit: AI Codebase Indexing 2026 – Zero-Setup Local Repo Mapping for Agents

使用 Dialectic Flow Financial Graph 分析市场趋势，模拟多方与空方辩论，获得清晰的投资洞察与报告。📊 Analyze market trends with the Dialectic Flow Financial Graph, simulating Bull and Bear debates for clear investment insights and reports.

以静态优先的学术写作平台，支持草稿撰写、引用管理、文献综述、图表与可投稿的 PDF 输出。Static-first academic writing platform for drafting, citations, literature review, figures, and submission-ready PDFs.

采用可复现研究方法，开发面向西部地震区的 AI 辅助地震预警系统。Develop AI-assisted earthquake early warning systems for Western seismic regions using reproducible research methods.

分配一个 AI 编码 agent 处理真实 issue，审阅其提出的方案，获取签名且可审计的 pull request。当前为单人模式，设计上支持扩展为多人协作模式。Assign an AI coding agent to a real issue, review the plan it proposes, and get a signed, auditable pull request back. Single-player today, designed to grow into shared, multiplayer use.

从 n8n 工作流向 Buzz 频道发送签名消息，将上下文、告警与输出接入 Agent 可直接执行的对话中。Send signed messages from n8n workflows to Buzz channels. Pipe context, alerts, and outputs into the conversations where your agents can action them.

基于 Torch2PC 的预测编码硕士论文可复现项目：在 Ubuntu/ROCm 上对 backpropagation 进行逐层与 compute-matched 对比，实现 PC-CATM/PC-TREF，对 state inference 进行机制诊断，并构建 QWake-PC 以实现自适应 exact inference —— 面向机制可解释的可复现预测编码研究。Воспроизводимый проект магистерской диссертации по predictive coding в Torch2PC: послойное и compute-matched сравнение с backpropagation, PC-CATM/PC-TREF, механизмная диагностика state inference и QWake-PC для адаптивного exact inference в Ubuntu/ROCm — reproducible research on mechanism-aware predictive coding.

一款全栈应用，通过 Chrome 扩展自动捕获求职申请，在统一面板中跟踪记录，并提供洞察以优化求职流程。A full-stack application that automatically captures job applications through a Chrome extension, tracks them in a centralized dashboard, and provides insights to improve your job search.

Research Paper Ranking Agent——用户向 Agent 提出研究问题，Agent 从 arXiv 检索学术论文，按相关性排序，并解释每篇论文排名的原因Research Paper Ranking Agent - a user gives the agent a research question, the agent retrieves academic papers from arXiv, ranks them according to relevance, and explains why each paper received its position.

基于 n8n 构建的多 Agent AI 研究助手——输入主题后自动规划研究问题、检索网络并生成完整报告；由 Groq (Llama 3.3) + Tavily 驱动Multi-agent AI research assistant built with n8n — takes a topic, plans research questions, searches the web and generates a full report automatically. Powered by Groq (Llama 3.3) + Tavily.

我的个人简历。My own Resume

拥有持久记忆的终端 AI 角色扮演客户端 —— 本地 SQLite、语义召回以及每次调用的成本统计。兼容 OpenAI，已在 OpenRouter 上测试。A terminal AI roleplay client with a memory that does not forget — local SQLite, semantic recall, and what every call cost. OpenAI-compatible, tested on OpenRouter.

分阶段的 AI 研究伙伴，辅助 UX 设计师从研究问题到独立可验证报告的完整流程。A staged AI research partner that takes a UX designer from research question to independently verified report.

PaperMind 是一个 AI 驱动的研究助手，可帮助用户撰写科研论文、开展文献综述、识别现有研究空白，运用多种机器学习模型与自然语言处理技术为用户提供洞察与建议The PaperMind project is an AI-powered research assistant that helps users in generating research papers, conducting literature reviews, and identifying gaps in existing research. It uses various machine learning models and natural language processing techniques to provide insights and suggestions to the users.

Kibin 是一家运营 kibin.com 的消费学习与学术写作公司，自 2011 年起受到学生信赖。其最新产品是一款 AI 驱动的学习伙伴（iOS 应用），可将讲座、笔记、PDF 和 YouTube 视频转化为个性化学习指南、测验、闪卡、转录稿与讲解，帮助学生更快学习……Kibin is a consumer study and academic-writing company operating at kibin.com, trusted by students since 2011. Its newest product is an AI-powered study partner (iOS app) that turns lectures, notes, PDFs, and YouTube videos into personalized study guides, quizzes, flashcards, transcripts, and explanations to help students learn faster and build…

Fintool 是一款面向对冲基金、资管机构及买方/卖方分析师等机构投资者的 AI 驱动股票研究分析师与金融副驾驶。其助手可摄取 SEC 文件（10-K、10-Q、8-K）、财报电话会议记录与金融数据，并附带来源引用地回答研究问题，加速……Fintool is an AI-powered equity research analyst and financial copilot built for institutional investors such as hedge funds, asset managers, and buy-side and sell-side analysts. Its assistant ingests SEC filings (10-K, 10-Q, 8-K), earnings-call transcripts, and financial data to answer research questions with source citations, accelerating…

用于 in-loop 设计审查的远程 MCP server，以及 OAuth 2.1 鉴权、SSRF 安全的 URL 处理、通过 MCP 执行长时任务的实践参考。A remote MCP server for in-loop design review, and a worked reference for OAuth 2.1 auth, SSRF-safe URL handling, and long-running jobs over MCP.

自包含的系统综述工作台——多库检索、双评审筛选、AI 辅助批判性评价（RoB 2、ROBINS-I、GRADE…）。单 HTML 文件，无后端，数据不离开浏览器。Self-contained systematic review workbench — multi-database search, dual-reviewer screening, AI-assisted critical appraisal (RoB 2, ROBINS-I, GRADE…). One HTML file, no backend, data never leaves your browser.

面向 MCP 服务器的混沌工程与韧性测试工具集A chaos-engineering and resilience testing toolkit for MCP servers.

交互式写作网站Interactive writing site

Narix Academy 的学术写作服务平台。Narix Academy's academic writing service platform

面向 AI 编程助手的 Agent skills —— 文档解析、PDF 工程、系统综述研究、闪卡与 WhatsApp 自动化。兼容 Claude Code、Cursor、Codex、ZCode。Agent skills for AI coding assistants — document parsing, PDF engineering, systematic-review research, flashcards, and WhatsApp automation. Works with Claude Code, Cursor, Codex, ZCode.

人工智能与软件开发公司；构建智能系统、自动化工具、定制软件及开源技术。شرکت هوش مصنوعی و توسعه نرم‌افزار؛ سازنده سیستم‌های هوشمند، ابزارهای اتوماسیون، نرم‌افزارهای سفارشی و فناوری‌های متن‌باز.

关于基于 AI Agent 的可靠企业数据智能（EDI）的研究笔记、研究日志、文献综述与 PoC 实验Study totes, research log, literature review, and PoC experiments on Reliable Enterprise Data Intelligence (EDI) with AI Agents.

通过分析 trace 诊断与调试 RAG 流水线，定位并修复影响答案质量与相关性的问题Diagnose and debug Retrieval Augmented Generation pipelines by analyzing traces to identify and fix issues affecting answer quality and relevance.

交互式矩阵：展示各类 UX 研究方法分别回答何种研究问题，覆盖 Discover/Design/Develop/Deploy 四个阶段。Interactive matrix: which UX research method answers which research question, across Discover/Design/Develop/Deploy.

AIMarket 的 10 款桌面与 IDE 应用 —— 基于 Flutter、Tauri 和 VS Code，统一托管于 Melos monorepo。10 desktop & IDE apps for AIMarket — Flutter, Tauri, and VS Code in one Melos monorepo.

10 门 AIMarket 学院实战课程 —— 编排、oracles、MCP 安全、Agent 经济（en/ru/es/fr/zh）。10 hands-on AIMarket academy courses — orchestration, oracles, MCP security, agent economy (en/ru/es/fr/zh).

🌐 在数据受限条件下重新思考多模态大语言模型的设计与扩展，以 NaViL 通过 Native Training 提升效率与性能🌐 Rethink Multimodal Large Language Models design and scaling under data constraints with NaViL, enhancing efficiency and performance through Native Training.