Lsong
Community@lsongdev · ✈ Beijing, China
@javascript & @golang; founder of @sczh-inc, former @bytedance @alibaba @meituan staff; creator of @kelpjs, @escpos and many other open-source projects.
Agent Skills by Lsong
Showing 61 vetted skills indexed across 1 GitHub repositories.
lsongdevwooyun
Provides WooYun vulnerability case data and statistics for business logic security testing.
lsongdevtwitter-downloader
Download Twitter/X tweet text, images, GIFs, and videos via fxtwitter and vxtwitter APIs.
lsongdevhyperframes
Create HTML-based video compositions with GSAP timelines, captions, transitions, and audio-reactive visuals.
lsongdevus-stock
Fetches real-time U.S. stock quotes from Google Finance via browser automation.
lsongdevasc-cli
Manages App Store Connect operations including TestFlight builds, releases, and reviews via the asc CLI.
lsongdevdeepseek-usage
Queries DeepSeek API balance, spending, and token usage via browser automation.
lsongdevhealth-sleep-analysis
Analyze Apple HealthKit sleep, SpO2, heart rate, and HRV data into visual reports.
lsongdevhyperframes-cli
Render HTML compositions into MP4 videos using the HyperFrames CLI and browser-based frame capture.
lsongdevmiya-desktop
Configure and troubleshoot Miya Desktop providers, profiles, agents, and runtime files.
lsongdevytmusic-hub
Manage YouTube Music playlists, liked songs, and search via ytmusicapi with cookie-based authentication.
lsongdevproduction-agent-public
Generates deployable code with ReAct reasoning, error handling, retries, and Docker deployment instructions.
lsongdevmemos-api
Generates curl and Python requests for the Memos REST API with authentication and pagination.
lsongdevweb-content-extractor
Extract clean main content from webpage URLs as Markdown using Defuddle or Jina AI reader services.
lsongdevtavily-search
Performs real-time web searches via the Tavily API and returns titles, URLs, and summaries.
lsongdevself-improving-agent
Log errors, corrections, and reusable learnings into structured Markdown records.
lsongdevcloudflare-dns
Manage Cloudflare DNS records via the Cloudflare API using curl commands.
lsongdevfucai3d-latest
Queries latest China Welfare Lottery 3D results and generates entertainment number recommendations.
lsongdevshortcut-share-file
Shares iCloud Drive files via the iOS Instant Share Shortcut URL scheme.
lsongdevswe-cli-skills
Provides expert CLI workflow guides covering 23 command-line tools across cloud, containers, and databases.
lsongdevqbt-hub
Manage qBittorrent downloads via the WebUI API using a Python CLI script.
lsongdevposthog-analytics
Automate PostHog dashboard creation, sync, update, and export via the PostHog API.
lsongdevbilibili
Query Bilibili videos, users, search, and trending content via CLI with YAML output.
lsongdevweb-design-guidelines
Review UI code against Vercel Web Interface Guidelines and report findings by file and line.
lsongdevglowmotion
Generate animated technical diagrams as self-contained HTML files with SVG, CSS, and SMIL.
Frequently Asked Questions About Lsong
FAQPage SchemaWhat tasks can I accomplish with Lsong's skill registry?▼
You can read and write Twitter/X, Telegram, YouTube Music, Spotify, Notion, Bilibili, Weibo, and Xiaohongshu data; fetch real-time stock quotes and fund flows; run reverse-DCF valuations; perform WooYun-based security testing; generate videos, images, TTS audio, PDFs, and presentations; and manage Cloudflare DNS, qBittorrent, and App Store Connect.
Who is the target audience for these skills?▼
Developers, security researchers, and power users who need programmatic access to Chinese and global platforms. Typical users include pentesters citing WooYun cases, traders checking A-share/US/KOSPI quotes, content creators rendering HyperFrames videos, and engineers managing TestFlight builds or Cloudflare DNS records.
What are the runtime requirements and dependencies?▼
Most skills require python3; several use browser_use for cookie-based authentication (Twitter/X, YouTube Music, Quark). Specific needs include curl for Memos, Telethon for Telegram, spotipy for Spotify, notion-client with NOTION_API_KEY, and DEEPSEEK_EMAIL/DEEPSEEK_PASSWORD env vars for DeepSeek usage checks.
Are these skills open-source and free to use?▼
Yes. Adapted skills like hyperframes and hyperframes-cli derive from HeyGen's Apache-2.0 originals with MIT-licensed adaptations; vercel-react-best-practices is MIT. Many data skills (stock-price, korea-stock, fucai3d-latest) use public endpoints requiring no API key or login.
How do authentication and credentials work across the platform skills?▼
Skills favor cookie or token-based auth: Twitter/X uses auth_token plus ct0 cookies fetched automatically via browser_use; Telegram requires one-time phone verification then persists sessions in SQLite; Notion uses a NOTION_API_KEY env var; Memos uses Bearer tokens stored in environment variables.