blogger-distiller

Distill Xiaohongshu and Douyin blogger playbooks into HTML reports and installable AI skills.

620|73|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/otter1101/blogger-distiller --skill blogger-distiller
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blogger-distiller
Source: https://github.com/otter1101/blogger-distiller
Command: npx skills add https://github.com/otter1101/blogger-distiller --skill blogger-distiller

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai-whisper, python-docx, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Manually studying a benchmark blogger's notes rarely yields a clear picture of why their content goes viral, and generic AI imitation prompts produce vague results. This Skill automates the full pipeline of collecting a blogger's public posts via the TikHub API, analyzing engagement patterns, and distilling their cognition, strategy, and content formulas into reusable outputs.

Core Features & Use Cases

  • Automated data collection: Fetches a blogger's profile, 30/50/80 posts, details, and comments from Xiaohongshu or Douyin through TikHub REST API endpoints with multi-endpoint fallback, rate limiting, and checkpoint recovery.
  • Three-layer distillation: Extracts the cognition layer (beliefs, viewpoints), strategy layer (posting rhythm, series planning), and content layer (title formulas, opening templates, CTA strategies) from real posts.
  • Dual deliverables: Generates a single-file HTML distillation report for humans and an installable creation-guide Skill folder (SKILL.md) that teaches your AI to write in that blogger's style.
  • Optional Whisper transcription: Converts video voiceovers into text transcripts to enrich the analysis with spoken content.
  • Use Case: Tell your AI assistant to distill a Xiaohongshu blogger you admire; 30 minutes later you receive an HTML report of their viral formulas plus a Skill folder you can load to generate new posts in their style.

Quick Start

Ask your AI assistant to distill a specific Xiaohongshu or Douyin blogger by name, then answer the prompts for platform, mode, and note count.

Frequently Asked Questions about blogger-distiller

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

FAQPage Schema
How do I analyze a Xiaohongshu blogger's content strategy?

Run the distillation flow by telling your AI assistant to distill the blogger by name. The tool collects 30, 50, or 80 public posts via the TikHub API, analyzes engagement statistics and title patterns, then generates an HTML report and a creation-guide Skill folder.

What API do I need to scrape Douyin or Xiaohongshu data?

This tool uses the TikHub REST API with a Bearer token. Register at user.tikhub.io, top up pay-as-you-go credit, enable all Xiaohongshu and Douyin endpoints in the console, and provide the token via environment variable or config file.

Does blogger-distiller work with both Xiaohongshu and Douyin?

Yes, it supports both platforms through separate crawl scripts routed by a --platform flag. You must explicitly state the platform before running, since the tool never infers it from the blogger's name.

Why do all TikHub API endpoints fail during collection?

Endpoint failures with HTTP 402 or 403 usually mean API permissions are not fully enabled or the balance is empty. Log into the TikHub console, enable all platform endpoints under API permissions, and verify your account balance.

Can I transcribe video voiceovers for deeper analysis?

Yes, install openai-whisper and ffmpeg to enable optional transcription of videos under 10 minutes. The default base model takes about 8-12 seconds per video, and transcripts are saved as Markdown files for the AI to analyze.

Is commenter privacy protected when collecting comments?

Yes, commenter identities are anonymized by default to Reader1, Reader2, or Author. Nicknames, user IDs, avatars, and IP locations are never saved, while comment text is retained as research material.