Douyin Daily Like Surge Leaderboard

Problem it solves

Short-video discovery is hard when growth happens inside a single day. Manually browsing feeds cannot reliably surface posts that gained the most likes in the last 24 hours. Douyin Daily Like Surge Leaderboard turns that into a daily data task: it ranks content by new likes added in one day and returns a reviewable, filterable TOP50 list. This makes it useful for tracking short-term hot topics, competitor content performance, and content ideas inside a vertical.

How it works

The skill calls an external ranking API and expects an API key in the X-API-KEY header. The normal flow starts with a date check: requests for “today” or “latest” are answered with yesterday’s data, future dates are redirected to yesterday, dates within the last 30 days are queried directly, and older dates are capped by the 30-day limit. Next, it maps the request to one of the supported verticals; when the category is ambiguous, it asks the user to choose. By default it returns yesterday’s all-category TOP20. If the user explicitly asks for the full list, it expands to up to 50 rows. The result is rendered as a Markdown table, with post titles as Markdown links to the original works, plus fields such as author, category, new likes, new favorites, new comments, new shares, and publish time.

Boundaries to keep in mind

This is a daily incremental view, not a real-time leaderboard, cumulative leaderboard, or hourly monitor. Data is updated at 17:00 for the previous day, supports up to 30 days of history, and returns at most 50 items per category. Because the numbers are daily interaction snapshots, spikes may be affected by posting cadence, paid boosts, or platform events. It should not be treated as a full Douyin content database.