Chinese Finance Writing De-AI Assistant

Problem

Chinese finance writing often has an obvious AI tone: report-style boilerplate, stacked parallel clauses, formulaic connectors, abstract jargon, and vague risk disclosures. Readers usually want to know why a stock looks attractive or risky, what the numbers imply, and where the real uncertainty is. This skill is an editor for existing Chinese financial and investment-research text, not a general-purpose copywriter or a source of investment advice.

How It Works

It activates when the user asks to remove AI traces, make text sound human, or explain research notes more plainly. It then scans the draft against built-in patterns and rewrites the weak sections. Core capabilities include:
- Detecting general Chinese AI writing patterns such as over-elevation, notable that, not only... but also..., stacked four-character phrases, excessive parentheses, and passive bureaucratic tone;
- Detecting finance-specific patterns such as rating boilerplate, templated forecasts, mechanical valuation comparisons, formatted risk lists, and policy-interpretation clichés;
- Preserving original meaning, tone, key facts, and numerical context while replacing vague claims with concrete explanations;
- Returning a rewritten version plus change notes that identify the matched pattern types;
- Generating a Word export when explicitly requested, with Microsoft YaHei for CJK text and Arial for Western text.

Boundaries

The skill is suitable for rewriting, polishing, and reducing mechanical wording. It should not be used as the sole basis for financial verification, compliance review, or investment decisions. If the source text lacks key assumptions, data, or causal paths, the skill can improve clarity but will not invent reliable analysis. For ratings, target prices, forecasts, and risk disclosures, retain professional judgment and necessary disclaimers.