Short Video Topic Research

Problem

Short-video planning often gets stuck when hot lists, competitor posts, and comment feedback are mixed into one vague narrative. This skill targets Douyin, Kuaishou, and XHS, and turns raw material into reviewable fields: topic heat, title samples, body text, publish time, engagement signals, comment themes, creator profiles, and recent posts. It is useful for building an evidence pool before scripting, rather than generating generic suggestions.

How It Works

• Hot-list observation: use douyin hot-search, kuaishou hot-search, and xhs hot-search to capture current ranking signals.
• Keyword research: use platform search --keyword to find topics, audiences, competitors, and hooks.
• Content evidence: fetch details, comments, and replies by content ID or URL to judge openings, user pain points, and reusable expressions.
• Creator research: search candidate accounts first, then inspect profiles, recent posts, and series evidence.
• Output structure: keep XHS, Douyin, and Kuaishou evidence separate; preserve XHS note_url and note_id as returned; after JSON responses, follow up on title structure, hook placement, and recurring comment pain points.

Boundaries

This is read-only. It does not log in, post, like, or comment. It uses SOCIALDATAX_API_KEY; if balance is insufficient, do not retry repeatedly—complete the returned charge step, then rerun the same command. Use pagination tokens exactly as returned; do not truncate, rewrite, or rebuild them. If results are sparse, relax keywords or add audience, scenario, or brand constraints.