LinkedIn Post Generator

AI-powered tool that transforms your drafts and raw thoughts into polished LinkedIn posts using your unique writing style.

Features

1. Post Generator

• Paste existing LinkedIn posts (1-6 examples) to learn your style
• Add raw drafts/ramblings separated by #
• Generate 1-5 polished posts maintaining your voice

2. Hook Scoring & Optimization

• Score hooks on: Scroll-Stopping, Curiosity, Emotion, Specificity, Brevity
• Generate 3 variations with angle (contrarian, story, question, etc.)
• Detailed scoring with improvement suggestions

3. Tone Analyzer

• Extract voice fingerprint from 3-10 past posts
• Analyze: Vulnerability, Humor, Formality, Story Ratio
• Identify hook style, signature phrases, emotional palette
• Visual radar chart
• Generate tone-matched posts

4. Voice to Draft

• Record audio (Streamlit audio_input) or upload files (mp3, wav, m4a)
• Transcribe via OpenAI Whisper API
• Auto-clean filler words (um, uh, like, basically, etc.)
• Structure into Hook/Body/CTA format
• Generate polished LinkedIn Post
• Save directly to drafts

5. Style Comparison

• Compare your writing style against competitors or industry leaders
• Identify gaps and opportunities for differentiation

6. Competitor Tracker

• Track competitor LinkedIn posts and engagement
• Analyze posting patterns and content strategies
• Identify content gaps with gap analysis
• Separate Streamlit app interface (competitor_tracker_app.py)

7. Services Layer

• Post Service: Orchestrates post generation workflow (tone analysis → post generation → hook scoring)
• Analysis Service: Combines tone analysis with style comparison across multiple creators
• Enables API integration for programmatic access

Quick Start

git clone https://github.com/Khadejah14/AutoPosts.git
cd AutoPosts
pip install -r requirements.txt

Create .env file:

OPENAI_API_KEY=sk-your-openai-api-key

Run the main app:

streamlit run app.py

Run the competitor tracker app:

streamlit run competitor_tracker_app.py

Requirements

• streamlit
• openai
• requests
• beautifulsoup4
• plotly
• python-dotenv
• pandas

Project Structure

AutoPosts/
├── app.py                      # Main Streamlit app
├── competitor_tracker_app.py   # Competitor tracker Streamlit app
├── config.py                   # Configuration settings
├── llm.py                      # LLM integration utilities
├── utils.py                    # Utility functions
├── core/                       # Core framework
│   ├── __init__.py
│   ├── base.py                 # Base classes
│   └── registry.py             # Module registry
├── features/                   # Feature modules (independent)
│   ├── __init__.py
│   ├── hook_scorer.py         # Hook scoring & variations
│   ├── post_generator.py      # Post generation logic
│   ├── style_comparison.py    # Style comparison tool
│   ├── tone_analyzer.py       # Voice fingerprint extraction
│   └── voice_to_draft.py      # Voice transcription module
├── services/                   # Service layer (orchestrates features)
│   ├── __init__.py
│   ├── post_service.py        # Post generation workflow
│   └── analysis_service.py    # Voice analysis & style comparison
├── competitor_tracker/         # Competitor tracking module
│   ├── __init__.py
│   ├── analyzer.py            # Content analysis
│   ├── database.py            # SQLite database
│   ├── gap_analyzer.py        # Content gap analysis
│   ├── scraper.py             # LinkedIn scraping
│   └── tracker.py             # Competitor tracking logic
├── versions/                   # Post version templates
│   ├── __init__.py
│   ├── v1_hook_focus.json
│   ├── v2_storytelling.json
│   └── v3_contrarian.json
├── data/                       # User data storage
│   └── data.json
├── .env                        # Environment variables
├── requirements.txt            # Dependencies
└── Dockerfile                  # Docker configuration