Bilibili Video Deep Analysis Expert

Problem

For Bilibili video analysis, raw metrics such as views, likes, and coins only describe the outcome; they do not explain why a video performed. Operators often need to inspect title hooks, bullet-comment sentiment, visible user demands, and the structural patterns worth reusing. bilibili-video-extractor turns these scattered judgments into a reviewable workflow instead of returning a single score.

How It Works

The skill focuses on breaking down Bilibili videos and extracting operational insight rather than listing fields:
- Video data extraction: organizes playback, engagement, and publication information as the analytical baseline.
- Bullet-comment sentiment analysis: identifies genuine user reactions, complaints, resonance points, and latent needs.
- Video information decomposition: highlights title structure, cover value proposition, content pacing, and reusable phrasing.
- Recommendation output: provides title formulas, topic directions, and content inspiration with an emphasis on actionable use.

When the full workflow is triggered, it follows references/core_workflow.md, requests fields step by step, and writes results into JSON. This makes it more suitable when a target video exists and a structured conclusion is needed.

Boundaries

It fits Bilibili content operations, competitor breakdowns, and topic research. It is not intended to invent conclusions without sufficient video context. Analysis quality depends on the provided video metadata, bullet-comment samples, and field completeness. For live scraping, licensed content, private content, or strict data-audit scenarios, verify data provenance and compliance first.