6_blog_post_generator_agent

Blog Post Generator Agent ✍️

A powerful multi-agent content creation pipeline that transforms a topic into a complete content suite: research, blog post, LinkedIn post, and Twitter thread.

Architecture

The system uses 4 specialized agents working in sequence:

1. 🔍 Research Agent

• Uses Exa to search for 5-7 high-quality, authoritative sources
• Extracts key facts, statistics, and expert insights
• Organizes findings into a structured research summary
• Output: Comprehensive research with citations

2. ✍️ Blog Writer Agent

• Transforms research data into engaging blog posts
• Creates SEO-optimized content with proper structure
• Includes headlines, sections, takeaways, and citations
• Output: Publication-ready blog post

3. 💼 LinkedIn Agent

• Converts blog content into professional LinkedIn posts
• Optimizes for LinkedIn's audience and algorithm
• Adds engagement hooks and hashtag suggestions
• Output: LinkedIn post (3-4 paragraphs + hashtags)

4. 🐦 Twitter Agent

• Creates viral-worthy Twitter threads from blog content
• Respects 280-character limit per tweet
• Adds hooks, insights, and calls-to-action
• Output: Twitter thread (3-5 tweets)

Pipeline Flow

Topic Input
    ↓
🔍 Research Agent (Exa Search)
    ↓
✍️ Blog Writer Agent
    ↓
    ├── 💼 LinkedIn Agent
    └── 🐦 Twitter Agent (parallel)
    ↓
Complete Content Suite

How to run

1. Install dependencies:

   uv pip install -r requirements.txt

   (Note: pyproject.toml is used, uv will handle it).

2. Set up your environment variables: Create a .env file in the root of the project and add your API keys:

   OPENROUTER_API_KEY="your-openrouter-api-key"
   OPENROUTER_MODEL_ID="minimax/minimax-m2:free"  # Optional, defaults to minimax/minimax-m2:free
   EXA_API_KEY="your-exa-api-key"
   

3. Run the Streamlit app:

   streamlit run 6_blog_post_generator_agent/streamlit_app.py