Skills
Core Technologies & Domain Expertise
generate-resume
Generate a tailored, ATS-optimised resume for a specific job posting. Input: a job URL and a candidate summary file (e.g. ZhangSan-summary.md). Output: a Markdown resume aligned to the posting with an appendix of unmet requirements. Supports single and batch modes. Invoke whenever the user wants to create a resume, generate a CV, tailor a resume to a job posting, apply for jobs, batch-generate resumes for multiple positions, or mentions anything about resumes, CVs, cover letters, or job applications — even casually like 'help me with my resume' or 'I need to apply for this job'.
find-git-leak
Scan git repos, directories, or files for leaked secrets (API keys, passwords, tokens) using gitleaks. Use this skill whenever the user wants to detect secrets, check for credential leaks, scan for sensitive data in code, audit a repo for exposed API keys or tokens, or run a security scan on their codebase. Also trigger when the user mentions 'secret scan', 'leak detection', 'credential scan', 'find leaked keys', 'check for hardcoded passwords', or any variation of detecting sensitive information in code or git history.
cloakbrowser
Use this skill for browser automation on websites protected by anti-bot, anti-scraping, or fingerprinting systems. Trigger this skill when the user needs to scrape protected websites, bypass Cloudflare Turnstile, reCAPTCHA, DataDome, Kasada, or Akamai, perform stealth browsing, manage multiple browser identities, solve captchas, spoof browser fingerprints, or automate interactions on sites that block standard playwright or puppeteer automation. Supports Playwright-compatible browser automation in both Python and Node.js (TypeScript/JavaScript). allowed-tools: Bash(python:*) Bash(python3:*) Bash(py:*) Bash(npm:*) Bash(npx:*)
literature-retrieval
Multi-source academic literature retrieval skill. Searches across 5 authoritative scholarly APIs (OpenAlex, Semantic Scholar, arXiv, CORE, Crossref) to find papers matching user-specified topics and filters. Generates a structured results report in Markdown. Use this skill whenever the user wants to search for academic papers, find research literature, do a literature search, retrieve scholarly articles, look up academic references, or gather papers on a topic — even if they don't explicitly say 'literature retrieval.' Also trigger on Chinese phrases like 文献检索, 论文检索, 搜索文献, 查找论文, 文献调研.
video-download
Download online videos and audio using yt-dlp. Use this skill whenever the user wants to download a video, extract audio from a URL, download YouTube/Bilibili/Twitter content, save a playlist, download subtitles, or grab any online media. Also use this skill when the user pastes a video URL and wants to save it locally. Works with thousands of websites including YouTube, Bilibili, Twitter/X, TikTok, and more.