skills.rest
TeamWiseFlow
TeamWiseFlowOFFICIAL
Explore all skills by this author →
GitHub

wx-channel-style-profiler

Extract 16-dimension style DNA reports from WeChat Channels video transcripts and aggregate them into production templates.

8.5k|1.4k|Updated Apr 24, 2024
Search Tags:#template generation#transcript analysis#style analysis#video script#short video#wechat channels#content dna
One-click install
npx skills add https://github.com/TeamWiseFlow/wiseflow --skill wx-channel-style-profiler
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: wx-channel-style-profiler
Source: https://github.com/TeamWiseFlow/wiseflow/tree/main/crews/main/skills/expert-wx-channel/tools/wx-channel-style-profiler
Command: npx skills add https://github.com/TeamWiseFlow/wiseflow --skill wx-channel-style-profiler

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Replicating the style of successful WeChat Channels (视频号) videos is usually guesswork. This Skill turns spoken scripts and transcripts into structured 16-dimension DNA reports, then aggregates multiple reports into a weighted DNA document and an executable production template so new videos consistently match a proven style.

Core Features & Use Cases

  • Single-video DNA report: Generate a 16-dimension analysis scaffold from a transcript (.md/.txt), with optional cover image, sample weight, focus dimensions, and source video reference.
  • DNA aggregation and template building: Merge all reports under a DNA ID with weighted statistics (weighted median, MAD, stable terms) into a DNA document and a five-stage production template (hook, empathy, credibility, value, closing).
  • Incremental updates and user-input translation: Add new samples or fuse user feedback (e.g. "move the conflict earlier") by mapping it to specific dimensions and revising the DNA document and template.
  • Use Case: Collect transcripts from ten benchmark WeChat Channels videos, run report on each, then run build to produce a DNA template that guides topic selection, titles, cover image AIGC prompts, and script structure for your own channel.

Quick Start

Ask the agent to run wx-channel-style-profiler report on a transcript file with a DNA ID and sample ID, then run build with that DNA ID to generate the aggregated DNA document and production template.

Frequently Asked Questions about wx-channel-style-profiler

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

FAQPage Schema
How do I analyze the style of a WeChat Channels video?▼

Run the report command with a transcript file (.md or .txt), a DNA ID, and a sample ID. The script generates a 16-dimension DNA report scaffold covering hook design, narration language, script structure, and more, which the agent then completes by re-reading the transcript.

How do I build a content production template from multiple videos?▼

Place multiple DNA reports under the same DNA ID and run the build command. It aggregates all reports with weighted statistics into a DNA document and derives a five-stage production template covering hook, empathy, credibility, value, and closing sections.

Can I analyze a video file directly without a transcript?▼

No. The tool only accepts text transcripts or spoken scripts in .md or .txt format. You must first obtain the text via ASR transcription, user-provided scripts, or analysis notes before running the report command.

What does the cover image analysis require?▼

The report command accepts a local .jpg, .jpeg, .png, .webp, or .gif cover image via --cover-image. A vision model must then read the image to fill in subject, composition, color, and AIGC reproduction prompt elements; without an image, that dimension is marked as not provided.

How do I update a DNA profile when I have a new video sample?▼

Use the update command with the new report file plus the existing DNA document and template paths. The script merges historical and new reports, recomputes weighted statistics, and preserves prior agent-written content for revision.

What are the limitations of the automated statistics?▼

The script only computes sentence length, punctuation density, person references, and two-character term frequencies as evidence candidates. It produces no scores or style verdicts; all 16-dimension conclusions must be made by the agent re-reading the source material.

Related Skills

Explore More Content & Communication Skills→
SpaceZephyrSpaceZephyr
333

baoyu-post-to-wechat

Publish markdown, HTML, or plain text to WeChat Official Account drafts and image-text posts.

Community
Advanced
aimentoraiaimentorai

content-craft

Generate TikTok titles, scripts, copy optimization, and safety reviews.

Community
Intermediate
perfectuser21perfectuser21

media-scraping

Scrape media performance data from six Chinese social platforms via Chrome CDP.

Community
Intermediate
mtsatryanmtsatryan
1

prompt-engineer

Design, optimize, and evaluate LLM prompts across RAG workflows and benchmarks.

Community
Advanced