WeChat Article Summary

Problem

WeChat articles, industry essays, and competitor docs often exceed the time a reader can spend on them. The opening may contain a thesis, the middle may hide key metrics, and the ending may include actionable advice, but signal density is uneven. Pasting the full text into a model can produce vague summaries, while feeding a link to a generic summarizer may drop critical numbers and executable conclusions. This skill targets the “too long to finish but necessary to extract” case by compressing input into a reviewable structured summary.

How It Works

It accepts an article link or pasted full text. First it confirms the minimum information set: input source, output format, and expected completeness. Then it identifies core viewpoints, key data, and action recommendations, and emits results in markdown, json, or document form. Uncertain items use a three-tier confidence model:
- ≥90%: output directly
- 85%–90%: mark “review recommended”
- below threshold: state that it cannot be confirmed and avoid guessing

Boundaries

It suits content operations topic triage, product manager competitor reading, student/researcher literature review, and readers who only need the core of a long article. If data is missing, context is incomplete, or the request exceeds the summarization boundary, it should say so explicitly. It does not provide legal, financial, tax, investment, or medical advice; for professional decisions, consult licensed experts and verify conclusions yourself.