Explore skills

Downloads PDFs from user-provided URLs to the default Downloads folder, then creates a same-name Markdown OCR copy beside the PDF. It supports direct PDF links, HubSpot tracking links, redirected download links, cloud-drive share links (Google Drive, Dropbox, SharePoint, OneDrive, Chinese), or webpages exposing PDF downloads.

Delegate code review, code implementation, or image generation to the OpenAI Codex CLI for a second opinion. Image generation works through ChatGPT login alone — no OPENAI_API_KEY required.

Provides guidance on GDPR Data Protection Impact Assessments (DPIA) under Article 35, incorporating EDPB guidelines, opinions on AI, and national SA blacklists/whitelists.

Automatically converts prototype code into static HTML long pages with intelligent annotation cards, supporting automatic requirement extraction and zero-loss restoration.

Collects Bilibili video comments for UP主 self-analysis, generating JSONL and summary.json for Claude Code analysis (sentiment, keywords, trends). Use to collect, download, export, or analyze comments on a Bilibili video (BV号 / URL / UID). It is read-only.

Create zero-context, executable specifications that can be reliably followed by humans or agents. This skill is used for writing, revising, evaluating, or operationalizing various types of specifications and procedural documents.

Codebase-aware spec interviewer. Analyzes the project, conducts a rigorous 1-on-1 interview with active pushback, validates technology choices against real-world adoption data, and produces a production-grade specification.

Generates optimized prompts for Google's Nano Banana (Imagen) image generators and creates brand-aligned visual content. Use this skill when uploading a reference image, requesting an image prompt, or using specific commands.

Bug Hunter v1.1.0 Systematically hunts and detects potential bugs in code, including security vulnerabilities, null safety issues, boundary conditions, exception handling gaps, logic defects, code smells, and concurrency problems.

Multi-agent parallel code review of staged git changes. Launches two independent reviewer agents (Security/Bugs + Architecture/Quality), runs a code simplification pass, cross-references findings, and produces a consolidated verdict.

This skill tracks weekly changes in Claude Code usage habits to identify improvements. It triggers on requests for weekly reviews, habit tracking, or checking for improvements over time, focusing on time-series analysis.

This skill fetches and displays Weibo hot topics. It allows querying hot topics, service status, and grouping titles. By default, it suggests creating a scheduled task and provides the cron command. It can be triggered by scheduled messages or the keywords "微博" and "微博热点".