Research & Data

This is intended as a showcase of my multidisciplinary skills and expertise in robotics, automation, AI/ML, and software. In this project, I leverage AI agents usage and workflows.

Enterprise-grade Codex prompt architect skill with plan-only, approval-gated execution, browser QA, security constraints, and prompt evaluation workflows.

Compare AdaL and Claude Code on Autoresearch benchmarks to find better hyperparameters, run more experiments, and converge faster

Enforce research mode in Claude Code with source citations, direct quotes, and grounded answers to cut hallucinations

Analyze Claude Code’s BUDDY pet system from leaked source, listing all 18 ASCII pets, render logic, and system details

Explore a public Claude Code source snapshot for security research, supply-chain analysis, and CLI architecture review

Enhance AI coding assistants to generate accurate SwiftData code, handling models, queries, predicates, migrations, and iCloud sync effectively.

Pi coding agent skill: scaffold and manage data broker removal apps (Incogni-like service)

面向 AI Agent 的百度 API 搜索 Skill，基于百度 AI Search / 百度百科 API，需百度 API Key

Arkon: Enterprise AI Knowledge Hub & MCP Server. Self-hosted knowledge base for teams to manage RAG contexts, access policies, and AI skills. Connect Claude and other LLMs via Model Context Protocol (MCP) for automated, secure organizational knowledge integration.

倪海厦中医课程资料的 Agent Skill：支持课程检索、方证穴位辨析、学习笔记整理与板书截图证据索引。 | An Agent Skill for Ni Haixia TCM course study, formula-pattern lookup, acupoint reference, and screenshot evidence indexing.

Provide AI agents with full Tor network access and dark web data through a zero-config OpenClaw skill or standalone tool.

Cross-source academic literature scan skill for Claude Code: fans out to OpenAlex / Semantic Scholar / arXiv / Perplexity, dedupes deterministically, reconciles into one cited consensus report.

Implement the Recursive Language Model pattern to manage large data externally and optimize code analysis in Claude Code and OpenCode environments.

考前期末速通 · Claude Code Skill · 资料速通 / 出题 / 解题 / 出卷，理科文科医学法学工科通用

Project Mental Model — a Claude Code skill that keeps a never-stale mental model of any codebase. Bundles the codegraph structure engine (PyPI: codegraph-pmm).

Multi-role plan verification skill for Claude Code — protocol-driven panel of architect/qa/security/judge + optional cross-model verify via GPT-5 and Gemini 2.5 Pro.

Claude Code skill: persona-based strategic product analysis — 5 users, 3 recommendation layers, elevation moves

周报 skill：从 Claude Code 会话自动生成日报/周报

Raptor turns Claude Code into a general-purpose AI offensive/defensive security agent. By using Claude.md and creating rules, sub-agents, and skills, and orchestrating security tool usage, we configure the agent for adversarial thinking, and perform research or attack/defense operations.

Skill library for Claude Code — opinionated rules and slash commands for modern ABAP (BTP ABAP Environment, S/4HANA 2023+ ABAP Cloud development model). Assumes the official SAP ABAP MCP Server provides system access.

Independent Agent/Codex skills for high-leverage data science workflows, with hero shots, infographics, validation scripts, and provider interoperability notes.

Description: Agent skill for working with Jinshuju forms, data, and workflow automation.

Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.