AI-Powered Blog Writing Assistant

An automated blog writing system that leverages CrewAI to create high-quality, well-researched blog posts. The project implements a multi-agent workflow for researching topics, generating content, and publishing blog posts with minimal human intervention.

Core Components

1. Research Crew

• Research Specialist: Gathers and processes information on the given topic
• Knowledge Base: Populates a Retrieval-Augmented Generation (RAG) system with relevant data

2. Writing Crew

• Draft Creator: Generates initial blog post drafts
• Senior Writer: Refines content, ensures quality, and adds relevant images

3. Tools & Integrations

• RAG System: For fact-based content generation
• Pexels API: For sourcing relevant cover images
• Dev.to Integration: Optional publishing capability
• Token Tracking: Monitors API usage and costs

Project Structure

• /crew: Contains agent implementations and configurations

  • config/: YAML files defining agent behaviors and tasks
  • blog_crew.py: Manages the blog writing agents
  • research_crew.py: Handles research-related agents

• /flow: Defines the workflow

  • blog_flow.py: Main workflow orchestration

• /output: Generated content storage

  • blogs/: Final blog posts
  • drafts/: Initial versions
  • research/: Collected research data
  • publication/: Publishing metadata

• /tools: Custom tools and integrations

  • rag_tool.py: Retrieval-Augmented Generation
  • image_search_tool.py: Image sourcing
  • devto_publisher_tool.py: Publishing interface

Workflow

1. Topic Research: System gathers and processes information
2. Draft Creation: Initial content generation
3. Content Refinement: Polishing and enhancement
4. Publishing: Optional deployment to platforms