Douyin Public Data Crawler

Problem

Short-video teams often need to collect public data around keywords, competitor accounts, comments, and trending lists, but manually copying pages and stitching spreadsheets is hard to scale for batch comparison or trend review. Douyin Smart Crawler targets this data-analysis workflow by turning scattered page data into reusable JSON logs.

How It Works

The skill is organized around four core actions: searching Douyin videos by keyword, filtering by likes or recency, batch-fetching public posts from a selected creator, and retrieving video comments and real-time trending data. After configuring GUAIKEI_API_TOKEN, choose the task-specific capability; outputs are usually saved under the skill's logs directory, which makes deduplication, field extraction, ranking comparison, and report generation easier to script.

• Topic analysis: combine keyword search with like-based ranking to identify patterns in high-engagement posts
• Competitor monitoring: batch-collect posts from benchmark accounts and compare cadence and content structure
• Sentiment review: read comment details to summarize emotions, requests, and friction points
• Trend tracking: observe trending-list changes to identify timely entry points

Boundaries

The skill handles public Douyin data only, not private or hidden content; a valid TOKEN is required; results are intended for personal or internal team analysis and should not be redistributed in violation of policy. If results are empty, check keywords, the --time range, and sorting parameters first.