wechat-style-profiler

Extracts 17-dimension style DNA reports from WeChat articles and aggregates them into writing templates.

8.5k|1.4k|Updated Apr 24, 2024
One-click install
npx skills add https://github.com/TeamWiseFlow/wiseflow --skill wechat-style-profiler
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wechat-style-profiler
Source: https://github.com/TeamWiseFlow/wiseflow/tree/main/crews/main/skills/expert-wx-mp/tools/wechat-style-profiler
Command: npx skills add https://github.com/TeamWiseFlow/wiseflow --skill wechat-style-profiler

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Replicating the writing style of successful WeChat Official Account articles is usually guesswork. This Skill turns style imitation into a repeatable process by extracting a 17-dimension DNA report from each article, aggregating reports into a DNA document, and deriving an executable writing template.

Core Features & Use Cases

  • Single-article DNA report: Analyzes one article across 17 dimensions (topic angle, title style, cover image, sentence rhythm, rhetoric, signature markers, and more), with weighted statistics and optional per-dimension focus.
  • DNA aggregation and template generation: Merges multiple reports by DNA ID with weights and focus scopes, producing a DNA document and a seven-part production template (topic, title, opening, development, turn, conclusion, CTA).
  • Incremental updates and user input translation: Merges new reports into existing DNA documents and translates free-form user instructions into concrete dimension-level writing rules.
  • Use Case: Collect five articles from a benchmark WeChat account, generate a DNA report for each, then build a DNA template so every new article you write matches that account's style, tone, and structure.

Quick Start

Ask the agent to run wechat-style-profiler report on a sample article with a DNA ID and sample ID, then run the build command to generate the aggregated DNA document and writing template.

Frequently Asked Questions about wechat-style-profiler

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze the writing style of a WeChat article?

Run the report command with the article file, a DNA ID, and a sample ID. The script computes sentence, paragraph, and punctuation statistics, then the agent fills in 17 dimension conclusions with evidence from the original text.

How do I create a writing template from multiple reference articles?

Generate a DNA report for each article under the same DNA ID, then run the build command. It aggregates all reports with their weights and focus settings into a DNA document and derives a seven-part production template covering topic, title, opening, development, turn, conclusion, and CTA.

What input formats does the style profiler accept?

The report command accepts .md and .txt article files; .docx files must be converted to text by the agent first. Cover images must be local .jpg, .jpeg, .png, .webp, or .gif files passed via the cover-image option.

Can I weight some sample articles more than others?

Yes. Pass a positive weight value per report and optionally restrict a report to specific dimensions using focus IDs. Aggregation uses weighted medians and distinguishes high-frequency commonalities from high-weight preferences and isolated cases.

What happens if no cover image is available for an article?

The cover image dimension is recorded as not provided and no visual evidence is fabricated. When a cover image is supplied, a vision model analyzes subject, composition, color, lighting, and style to produce AIGC prompt elements for reproducing the cover style.