Daily Hot News Summary Assistant

Problem

Opening multiple platform hot lists creates noisy duplication: the same story may appear on Weibo, Zhihu, and Bilibili, while high heat on a single platform does not always signal importance. This skill turns scattered rankings into a reviewable daily digest: it filters the most notable topics, explains why they matter, and identifies what to watch next.

How It Works

• Multi-source fetch and fallback: pulls hot lists from Weibo, Zhihu, Bilibili, Douyin, Baidu, 36Kr, Sspai, and Tieba, with fallback paths such as aggregator data or local cache, while preserving platform status.
• Structured scoring: ranks topics using cross-platform resonance, absolute heat, topic diversity, and platform balance to reduce single-platform dominance and duplicate items.
• Three-layer insight: each selected item includes What, So What, and Now What, covering the event essence, impact chain, and follow-up observation points.
• Style and format options: can switch between serious analysis, sharp commentary, investment view, and casual gossip, with output suited to chat groups, knowledge bases, dashboards, or voice briefings.
• Historical archive: stores data by day, enabling trend comparisons such as versus-yesterday or weekly-review analysis.

Boundaries

Hot-list data is time-sensitive, so fetch on demand and keep fetch frequency in mind. Platform status is marked as normal, degraded, or failed; treat that metadata as part of data reliability. Content filtering should still avoid sensitive topics, and scoring is better used as a prioritization signal rather than a substitute for independent fact verification.