仓库/Skill 库

基于 IQ 信号、采用 STFT 频谱图、视觉模型、k-fold 训练与集成推理的多节点 RF 无人机识别可复现研究工具包。A reproducible research toolkit for multi-node RF drone identification from IQ signals using STFT spectrograms, vision models, k-fold training, and ensemble inference.

2026 自主 AI Agent 平台 🚀——基于 MCP 与 DAG 的多 Agent SaaSAutonomous AI Agent Platform 2026 🚀 - Multi-Agent SaaS with MCP & DAG

🚀 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.

大语言模型监督微调的反馈对齐方法Feedback alignment for supervised fine-tuning of large language models

一个工作流框架，适配所有编码 agent。通过与工具无关的核心，将你的 AI 编码规范安装到 Claude Code、Cursor、GitHub Copilot、Gemini CLI、Codex 和 Windsurf。Apache 2.0。One workflow framework, every coding agent. Install your AI-coding discipline into Claude Code, Cursor, GitHub Copilot, Gemini CLI, Codex, and Windsurf from one tool-agnostic core. Apache 2.0.

von der Heyde, L., Keusch, F., Buskirk, T. D., & Eck, A. (2026). AI in the Loop?! A Systematic Review of the Use of Large Language Models in Survey and Public Opinion Research. SocArXiv. https://doi.org/10.31235/osf.io/eubj4_v1 的复现材料与文献数据库Replication materials and literature databases for von der Heyde, L., Keusch, F., Buskirk, T. D., & Eck, A. (2026). AI in the Loop?! A Systematic Review of the Use of Large Language Models in Survey and Public Opinion Research. SocArXiv. https://doi.org/10.31235/osf.io/eubj4_v1.

metaScreener——基于插件的桌面应用，用于 human-in-the-loop systematic literature screening。在顺序可审计的 pipeline 中结合确定性启发式过滤与 LLM 推理，通过 SHA-256 校验包实现完全可复现。MIT 协议。metaScreener — a plugin-based desktop application for human-in-the-loop systematic literature screening. Combines deterministic heuristic filters with LLM inference in a sequential, auditable pipeline. SHA-256 verified bundles for full reproducibility. MIT licensed.

Give pure-text LLM agents eyes: zero-dependency image understanding, OCR & image Q&A via MCP server + standalone Python script. 给纯文本 AI Agent 的图片理解能力(看图/OCR/图片问答)。

用于文献与系统综述的 AI agent 流水线，对齐 PRISMA 2020、Cochrane 与 GRADE。默认无需密钥（OpenAlex/CrossRef + 标准库）；在 LLM 薄弱处设人工把关。An AI-agent pipeline for literature and systematic reviews, aligned with PRISMA 2020, Cochrane, and GRADE. Keyless by default (OpenAlex/CrossRef + stdlib); human-gated where LLMs are weak.

💳 使用 XGBoost 检测金融交易欺诈，结合针对不平衡数据集与复杂特征的高级优化技术💳 Detect fraudulent financial transactions using XGBoost. Optimize performance with advanced techniques for imbalanced datasets and complex features.

面向 LLM 的阶段感知上下文窗口治理框架。提供不变的上下文长度上限、基于熵的稳定性控制，以及针对降级（碎片化）状态的概率性保证，适用于生产级 LLM 系统。Phase-aware context window governance framework for Large Language Models (LLMs). Provides invariant context length caps, entropy-based stability control, and probabilistic guarantees against degraded (fragmentation) states for production LLM systems.

这个仓库展示了我的技能、项目，以及从数据工程转型为数据科学家角色的持续学习历程。This repository showcases my skills, projects, and continuous learning journey as I transition from Data Engineering to a Data Scientist role.

硕士论文（KCL）：在攻击者-机器留出划分下重新衡量 IoT 入侵检测，并测试基于 LLM 的报文预测是否能提升性能。移除泄露后，macro-F1 从 0.90 降至 0.60。MSc dissertation (KCL): re-measuring IoT intrusion detection under an attacker-machine holdout, and testing whether LLM-based packet prediction improves it. macro-F1 0.90 -> 0.60 once leakage is removed.

🎨 利用 GPT、Gemini 等模型，AI 驱动的学术图表一键生成与定制工具🎨 Generate academic diagrams effortlessly with this AI-driven tool, leveraging models like GPT and Gemini for seamless creation and customization.

🛠 通过此硬件插件提升 vLLM 在 Kunlun XPU 上的性能，无缝集成主流 AI 模型并优化执行效率🛠 Enhance vLLM performance on Kunlun XPU with this hardware plugin, offering seamless integration for popular AI models and optimized execution.

🧪 实验性本地优先 RAG，附带桌面 GUI：将文档拖入文件夹即可对话聊天，内置 Ollama / fastembed / OCR / reranker。🧪 Experimental local-first RAG with desktop GUI — drop documents in a folder, chat with them. Ollama / fastembed / OCR / reranker built in.

使用部署在 streamlit 上的 MCP server 构建的旅行 Agent。A Travel Agent Built using a MCP server which is deployed in streamlit

基于 LLM 的自优化 markdown 维基构建工具，面向科研场景A self-refining LLM-powered markdown wiki builder for scientific research

🔍 Self-Corrective RAG 基于 LangGraph 与 Google Gemini 优化查询并评估文档相关性，提升检索效果🔍 Enhance your searches with Self-Corrective RAG, a system that optimizes queries and evaluates document relevance using LangGraph and Google Gemini.

自托管的个人上下文服务器 — 你与 AI Agent 共享统一的知识与信任边界：知识库、检索、凭证与 Agent 记忆，通过 MCP 挂载Self-hosted personal context server — one knowledge and trust boundary shared by you and your AI agents: knowledge base, retrieval, credentials, and agent memory, mounted over MCP

应用型 AI 网络安全研究实验室，涵盖机器学习威胁检测、对抗机器学习、鲁棒性、网络威胁情报、MITRE ATT&CK/ATLAS、LLM 安全与可复现研究。Applied AI Cybersecurity research labs covering machine learning threat detection, adversarial ML, robustness, cyber threat intelligence, MITRE ATT&CK/ATLAS, LLM security and reproducible research.

一款本地化、隐私优先的 LLM 助手，提供学术写作辅助、音频转写、文档检索与数据分析。A local, privacy-first LLM assistant that performs academic writing assistance, audio transcription, document retrieval, and data analysis.

为 AI agent 提供持久化、结构化记忆。基于 Go 的本地优先 MCP server：语义召回、去重、provenance、冲突解决。单一静态二进制，零依赖。Persistent, structured memory for AI agents. Local-first MCP server in Go: semantic recall, dedupe, provenance, conflict resolution. One static binary, zero dependencies.

SAM 的纯文档概述。SAM 是一个由 9 个 agent 组成的 LLM pipeline，用于自动化系统综述与 meta 分析。实现位于私有仓库。Documentation-only overview of SAM, a 9-agent LLM pipeline that automates systematic reviews and meta-analyses. Implementation lives in a private repository.

从零开始在 PyTorch 中构建 decoder-only Transformer，涵盖分词、预训练、监督微调，以及基于核心张量操作的对齐。Build decoder-only Transformers from scratch in PyTorch, covering tokenization, pretraining, supervised fine-tuning, and alignment using core tensor operations.

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"

面向科学与医学写作的人性化润色的阿英双语 skill，同时保持学术准确性。Bilingual Arabic-English skill for humanizing scientific and medical writing while preserving scholarly accuracy.

为仿制药临床法规事务团队打造的法规准备加速器。A regulatory prep accelerator for a generic-drug Clinical Regulatory Affairs team.

证据优先的科研桌面工具，支持论文发现、PDF 证据抽取、论点分析、对比与文献综述综合Evidence-first research desk for paper discovery, PDF evidence extraction, claim analysis, comparison, and literature review synthesis.

将任何内容转化为 token 优化、布局感知的 Markdown，并精确知道为你节省了多少。无损压缩、诚实的 token 报告，CLI + MCP 服务器 + agent skill。Turn anything into token-optimized, layout-aware Markdown - and know exactly what it saved you. Lossless compression, honest token reports, CLI + MCP server + agent skill.

使用 Qwen3-TTS 在本地 GPU 上克隆声音并从文本生成语音，提供端到端训练流水线。Clone voices and generate speech from text locally on your GPU using Qwen3-TTS with an end-to-end training pipeline.

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).

Conversational academic writing Copilot: LangGraph ReAct agent with citation tracing, faithfulness reviewer, RAG & Reflexion LaTeX self-repair ｜ 对话式学术写作 Copilot：LangGraph ReAct 智能体，引用溯源 + 忠实度审校 + RAG 检索 + Reflexion 编译自修复，端到端产出真实 PDF

保留图表的转换器 —— DOCX/XLSX 转 Markdown，采用原生 OOXML 图表数据提取（无需光栅化/OCR/VLM），并提供零损耗的复合图表标记。Converters where figures survive — DOCX/XLSX to Markdown with native OOXML chart-data extraction (no rasterize/OCR/VLM) and zero-loss composite-figure markers

集成 RAG、知识图谱、语义搜索、论文推荐与本地 LLMs 的 AI 研究助手AI Research Assistant with RAG, knowledge graphs, semantic search, paper recommendations, and local LLMs.

🌐 使用 LangExtract 无缝从文本中提取语言，简化语言检测，为项目提供便捷且高准确率的增强。🌐 Extract languages from text seamlessly using LangExtract. Simplify language detection and enhance your projects with ease and accuracy.