RAG parameter sweep and evaluation toolkit — a practical way to test how different vector databases, embedding models, chunking strategies, and retrieval methods perform across your own RAG setup. Built to help you move from guesswork to evidence, with support for any vector database.

A comprehensive Pre-RAG prototype dashboard for document parsing, multi-method chunking, vector embedding generation, and hybrid search exploration. Built with React and powered by in-browser ML models.

AI Agent skill for Zotero — search your local library with zero dependencies, no MCP server, no embedding

OpenClaw plugin for Google Workspace — Gmail (send/read/label/trash), Calendar (events + delete), Drive (list/share/trash), Docs (create/read/append), Sheets (create/read/append), Slides (create/read) via direct OAuth. Your OAuth client talks straight to Google. No MCP server, no third-party gateway, no IMAP App Password.

Drive Blender headless from your AI agent: scaffold, build, animate from natural language. JSON in, JSON out.

高德地图 Agent Skill（无需 MCP）— geocoding/POI/weather/route planning for Claude Code · pi · any shell agent, pure curl. AMap (Gaode) maps, no MCP server.

Azure DevOps PR and security review toolkit for GitHub Copilot, with reusable prompts and Go-based skills for secure, evidence-driven reviews.