Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →tsingyuai/media-crawler-zhihu
Collect Zhihu answers, articles, videos, comments, and creator evidence with MediaCrawler through search and exact URLs.
Passed 18 of 22 test cases with the skill, and 17 of 22 without it.
2 fixed16 passed either way3 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
When compiling the final summary report for a completed Zhihu data collection job on synthetic biology, is it sufficient to list only the search keywords, output directory paths, and total count of items?
| Checked | Without skill | With skill |
|---|---|---|
| Includes the system git commit hash in the summary report. | Missed | Passed |
| Includes filtered item exclusions in the summary report. | Passed | Passed |
The instructions your model reads when the skill loads.
Follow media-crawler. Require purpose, keyword/content/creator inputs, sample and date bounds, comments/media scope, and invoking Model Memory destination.
Search with --platform zhihu --type search using problem wording, concept/category and audience/use-case variants. Shortlist by relevance, argument/content-type diversity, recency, author fit and engagement. Configure ZHIHU_SPECIFIED_ID_LIST with full answer, article, or video URLs; configure ZHIHU_CREATOR_URL_LIST with full people URLs. Run detail/creator mode.
Preserve content type. For answers retain question context; for articles retain article title; for video identify unavailable transcript rather than inventing one. Separate author text from comments, deduplicate by canonical content ID, and retain discovery-query provenance. Enable comments/media only for a shortlist and stop on challenge/risk control.
Return content-type breakdown, inputs, raw/copied paths, detail/comment/media coverage, time, commit, exclusions and selection rationale.
test cases passed, vs 17 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.