SuperSearch Pro
Paste the following prompt into your AI chat to install this skill:
Please install @user_6455574a/super-search-pro by following https://skillhub.cn/install/skillhub.md.
About this skill
Problem
Cross-platform information is scattered across search engines, WeChat, Weibo, Zhihu, Bilibili, Douyin, Xiaohongshu, and other sites. Manual searches can produce duplicate results, mixed time signals, and inconsistent relevance. This skill turns one query into parallel searches and consolidates the fragments into a comparable structured report, useful for public-opinion tracking, technical research, policy review, and financial information triage.
How It Works and Limits
Core workflow:
- Parallel retrieval: starts with required platforms, then adds academic, financial, technical, or government sources by scenario; uses WebSearch indirectly for platforms with aggressive anti-scraping behavior.
- Normalization: converts results into JSONL and extracts relative time signals from snippets when available to support freshness scoring.
- Deduplication and ranking: uses MinHash + LSH for lower-cost deduplication, then combines TF-IDF, cosine similarity, keyword density, and time decay to produce a Top 10 list.
- Report output: generates .md, .json, and .checkpoint.jsonl for viewing, export, and resumable reruns.
Caveat: output depends on WebSearch coverage and the Python analysis environment; closed platforms may not expose full content. It is suited for fast aggregation and lead gathering, not a replacement for formal crawling or deep factual verification.
Use Cases
- Track a product launch discussion across platforms for one week and extract Top 10 relevant posts with source distribution.
- Research a technology framework by aggregating public discussions from GitHub, CSDN, and Juejin, then compare similarity.
- Compile policy-related reporting and platform commentary after a release, producing a Markdown report with freshness labels.
- Audit brand mentions on Weibo, Douyin, and Xiaohongshu for a negative public-opinion event and label each source platform.
Best For
- Public-opinion analysts who need to summarize cross-platform discussion and heat for an ongoing industry topic.
- Engineers doing technical selection who need to aggregate public tutorials, discussions, and case studies and rank relevance.
- Communications staff handling brand PR who need to check topic mentions across social and content platforms.
- Financial researchers who need to collect and rank relevant items from finance media and social platforms.
Related Skills
Search Huawei Cloud official docs and product pages to find ECS, OBS, RDS, CCE product specs, parameters, documentation, and API references without login.
OCR-based recognition for movie, train, flight, and event tickets in images or PDFs, extracting key fields into Markdown or JSON reports.
Turns notes, research, and meeting summaries into actionable next moves, plans, decisions, experiments, and decision-changing gaps.
A local wiki knowledge base manager that compiles raw documents into sourced, indexed Markdown pages with wikilinks, query support, and health checks.