Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →tsingyuai/media-crawler-bilibili
Collect Bilibili video and creator evidence with MediaCrawler through search, exact BV detail, comments, dynamics, contacts, and optional media.
Passed 18 of 23 test cases with the skill, and 10 of 23 without it.
and 13 more in the full report
9 fixed11 passed either way5 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
I need to configure date filtering in `config/bilibili_config.py` for Bilibili videos published between 2024-01-01 and 2024-01-31. Which configuration variable names should be assigned?
| Checked | Without skill | With skill |
|---|---|---|
| Sets START_DAY in config/bilibili_config.py. | Missed | Passed |
| Sets END_DAY in config/bilibili_config.py. | Missed | Passed |
The instructions your model reads when the skill loads.
Follow media-crawler. Define purpose, query/BV/creator inputs, date window, sample target, desired quality, comments/media scope, and invoking Model Memory destination.
Search with --platform bili --type search. Configure START_DAY, END_DAY, BILI_SEARCH_MODE, and conservative MAX_NOTES_PER_DAY in config/bilibili_config.py. Shortlist by relevance, content format, recency, creator fit and engagement—not raw plays alone. Put full video URLs or BV numbers in BILI_SPECIFIED_ID_LIST; put Space URLs or UIDs in BILI_CREATOR_ID_LIST. Use detail/creator mode respectively.
If downloading video, explicitly select BILI_QN compatible with account/video access and disclose that requested quality may be unavailable. CREATOR_MODE, contacts and dynamics limits are separate from content comments; record which were enabled. Deduplicate by BV, preserve query provenance, verify files, and stop on auth/risk control.
Return inputs, config window/mode/quality, raw/copied paths, counts and coverage, time, commit, partial failures, and selection rationale.
test cases passed, vs 10 / 23 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.