BlogWriter_Agent

This repository hosts BlogWriter_Agent, a multi-agent system for collaborative blog content generation using the CrewAI and Ollama frameworks. The system is designed to facilitate the creation of high-quality blog content through autonomous agents that handle content planning, writing, and editing in a coordinated workflow.

Table of Contents

• Introduction
• Features
• Requirements
• Installation
• Usage
• Configuration
• Agents

Introduction

BlogWriter_Agent automates the blog writing process using three specialized AI agents:

1. Content Planner: Plans the blog structure and gathers relevant information.
2. Content Writer: Writes content based on the planner's structure and guidelines.
3. Content Editor: Refines and edits the content to match stylistic and grammatical standards.

This project leverages the CrewAI framework to coordinate agent tasks and Ollama for local LLM management, with agents set up to perform tasks autonomously or in a collaborative fashion.

Features

• Multi-agent workflow: Allows agents to work collaboratively, each with specialized roles.
• LLM integration: Easily switchable models using Ollama’s local model setup.
• Configurable parameters: Define agent roles, goals, and objectives with flexibility.
• Privacy-focused: Runs models locally, so your data never leaves your machine.

Requirements

• Python 3.8+
• Ollama for local LLM management
• CrewAI for multi-agent orchestration

Installation

1. Clone the repository:

   git clone https://github.com/srishrachamalla7/BlogWriter_Agent.git
   cd BlogWriter_Agent

2. Install dependencies:

   pip install crewai==0.28.8 crewai_tools==0.1.6 langchain_community==0.0.29 

   OR

    pip install -r requirements.txt

3. Install Ollama:

   • Download and install the Ollama app for your operating system from Ollama's website.
   • Set the default path for storing downloaded models or configure the path using environment variables if needed.

4. Download the Llama 3 model:

   ollama run llama3

Usage

1. Set up the LLM with Ollama:

   • Create a ModelFile in your project directory with the following setup:

     FROM llama3
     
     PARAMETER temperature 0.3
     PARAMETER stop Result
     SYSTEM ""

   • Run the following command to set up the model in Ollama:

     ollama create crewai-llama3 -f .\ModelFile

2. Configure environment variables:

   • Make your OpenAI API Key:

     export OPENAI_API_KEY= 'NA'

3. Run the script:

   • The agents are set up in main.py with their respective roles. Execute the following to start the agents:

     python main.py

Configuration

where you can modify:

• LLM model: Set to use Meta Llama 3 by default.
• Agent parameters: Roles, goals, maximum iteration count, verbose logging, and more.
• Tasks: Define tasks to be executed by each agent.

Agents

Content Planner Agent

The Content Planner agent generates an outline for the blog post, including relevant topics and subtopics.

• Role: Content Planner
• Goal: Generate an informative, structured content outline.
• Parameters: Delegation disabled, verbose logging enabled.

Content Writer Agent

The Content Writer uses the planner’s outline to create a comprehensive draft with factual accuracy and depth.

• Role: Content Writer
• Goal: Write high-quality content based on the outline provided by the planner.
• Parameters: Delegation disabled, verbose logging enabled.

Content Editor Agent

The Content Editor refines and polishes the draft to ensure it meets professional standards and fits the publishing platform’s style guidelines.

• Role: Editor
• Goal: Edit the content for style, grammar, and adherence to journalistic standards.
• Parameters: Delegation disabled, verbose logging enabled.

---