技能列表

Write clean, maintainable pytest tests using Fake-based testing, contract testing, and dependency injection patterns. Use when setting up test suites for Python/MCP projects, creating Fakes for external dependencies, writing contract tests, or implementing test patterns with fixtures and parametrization.

Use Document Mind (DocMind) via Node.js SDK to submit document parsing jobs and poll results. Designed for Claude Code/Codex document understanding workflows.

|

Compare two versions of an OpenClaw skill to detect security-relevant changes. Use before updating any skill from ClawHub. Highlights new capabilities, changed patterns, and recommends whether an update is safe.

>

ONVIF PTZ control + RTSP capture + camera discovery. Works with any ONVIF Profile S/T camera. Tested with TP-Link, Hikvision, Dahua, Reolink, Amcrest, Axis.

Set up and operate a personal knowledge system using Supabase (pgvector) and OpenRouter. Five structured tables — thoughts (inbox log), people, projects, ideas, admin — with AI-powered classification, confidence-based routing, and semantic search across all categories. Captures thoughts from any source, classifies them via LLM, routes them to the right table (the Sorter), rejects low-confidence classifications (the Bouncer), and logs everything (the Receipt). Two opinionated primitives — Supabas

Korean website specialized scraper with anti-bot protection (Naver, Coupang, Daum, Instagram)

Get the real-time location (latitude/longitude) of the International Space Station.

Automatische Selbst-Verbesserung durch Fehler-Lernen und Pattern-Erkennung

Fetch OpenStreetMap vector data (streets, buildings) for an address and export to SVG, GeoPackage, or DXF for CAD/Rhino.

Systematically improve code through structured analysis-mutation-evaluation loops. Adapted from ALMA (Automated meta-Learning of Memory designs for Agentic systems). Use when iterating on code quality, optimizing implementations, debugging persistent issues, or evolving a design through multiple improvement cycles. Replaces ad-hoc "try and fix" with disciplined reflection, variant tracking, and principled selection of what to change next.