What problem does it solve?
Manually querying the TikHub YouTube API and wrangling its large raw JSON responses is repetitive and error-prone. This Skill provides minimal curl and Python request patterns, a raw-JSON-first saving convention, and a stdlib-only post-processor that converts responses into CSV and simplified JSON.
Core Features & Use Cases
- Single-request workflows: Run one YouTube general search or one video-detail lookup via curl or a tiny httpx snippet, with the full raw response saved for audit and replay.
- Raw JSON post-processing: The included postprocess_youtube_raw.py script reads one file or a directory of raw responses and writes youtube_search_summary.csv and youtube_search_summary.json.
- Pagination and batch guidance: Supports continuation-token pagination and low-concurrency batch requests for many keywords or video IDs.
- Use Case: A researcher needs structured metadata for 50 YouTube videos matching a keyword. They run one search request, save the raw JSON, then run the post-processor to get a clean CSV of titles, authors, view counts, and URLs.
Quick Start
Set the TIKHUB_API_KEY environment variable and ask the agent to search YouTube for a keyword through TikHub and save the raw JSON results.