This project addresses the challenge of understanding and replicating successful online content creators' strategies by automating the "distillation" of their cognitive frameworks, operational tactics, and content styles. It targets content creators, marketers, and AI users, offering a way to gain deep, actionable insights and integrate them into AI-powered content generation workflows.

How It Works

The system employs a hybrid "script + AI" architecture. Python scripts handle data acquisition via the TikHub API, statistical analysis, and pattern recognition, while AI models perform higher-level cognitive extraction, strategic analysis, and personalized recommendation generation. It distills insights into three layers: Cognitive (beliefs, viewpoints), Strategy (operations, timing), and Content (titles, hooks, CTAs), producing outputs for both AI agents and human review.

Quick Start & Requirements

• Installation: Clone the repository (git clone https://github.com/otter1101/blogger-distiller.git), navigate into the directory (cd blogger-distiller), and run the setup script (python install.py).
• Prerequisites: Python 3.10+ is required. Crucially, it necessitates API access and payment to TikHub (https://user.tikhub.io) for data retrieval, involving account registration, topping up, enabling API permissions, and generating a token.
• Links: Project GitHub, TikHub Registration.

Highlighted Details

• Dual Output: Generates an AI-ready "Creation Guide Skill" folder for direct integration into AI tools and a comprehensive HTML Distillation Report for human analysis.
• Deep Distillation: Extracts insights across Cognitive, Strategy, and Content layers, moving beyond surface-level metrics to understand core beliefs, operational rhythms, and specific content formulas.
• Extended Analysis: Offers advanced features like "Analyze Cover" (visual style breakdown) and "Keyword Trend Insight" (topic opportunity discovery) without requiring re-collection of data.
• Douyin Transcription: Integrates Whisper for video audio transcription, providing word-for-word transcripts for AI analysis (Base model ~70% accuracy).

Maintenance & Community

• Community: A WeChat group is available for discussion (vx:catsanddogs666).
• Development: Version history indicates active development and feature expansion, including multi-platform support and advanced analytical modules.

Licensing & Compatibility

• License: MIT License.
• Restrictions: Primarily intended for learning and research. Commercial use requires users to independently verify compliance with platform policies and TikHub terms. Prohibits selling/distributing data or using scraped data as a core commercial product.

Limitations & Caveats

• External Dependency: Relies heavily on the TikHub API, which incurs costs and is subject to its availability and terms.
• Commercial Use: Not explicitly endorsed; users must ensure compliance with all relevant laws and platform agreements.
• Audio Transcription: Douyin video transcription is limited to videos under 10 minutes, with a base model accuracy of ~70%; Xiaohongshu is not supported.
• Risk: Users assume all risks, including potential account issues, data accuracy, and API costs.