Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/tonykipkemboi/crewai-content-creator-flow


https://github.com/tonykipkemboi/crewai-content-creator-flow

Last synced: 3 days ago
JSON representation

Awesome Lists containing this project

README 

          
# Content Creator Flow - AI-Powered Content Generation System



A sophisticated multi-agent AI system powered by [crewAI](https://crewai.com) that generates high-quality content across different formats. This project demonstrates advanced flow-based orchestration of specialized AI crews for research reports, blog posts, and LinkedIn content creation.



## 🏗️ Architecture Overview



```mermaid

flowchart TB

    Start([User Input]) --> Router{Content Type
Router}

    

    Router -->|"1. Blog Post"| BloggerCrew[Blogger Crew]

    Router -->|"2. LinkedIn Post"| LinkedInCrew[LinkedIn Post
Generator Crew]

    Router -->|"3. Research Report"| ResearcherCrew[Researcher Crew]

    

    %% Blogger Crew Details

    BloggerCrew --> BA1[Senior Investigative
Technology Journalist]

    BA1 --> BA2[Senior Staff Writer
Technology Section]

    BA2 --> BA3[Editorial Director
& Senior Editor]

    BA3 --> BlogOutput[blog_post.md]

    

    %% LinkedIn Crew Details

    LinkedInCrew --> LA1[Senior LinkedIn
Post Researcher]

    LA1 --> LA2[Senior LinkedIn
Post Writer]

    LA2 --> LinkedInOutput[linkedin_post.md]

    

    %% Researcher Crew Details

    ResearcherCrew --> RA1[Senior Content
Researcher]

    RA1 --> RA2[Senior Content
Report Generator]

    RA2 --> ResearchOutput[report.md]

    

    %% Styling

    classDef crew fill:#e1f5fe,stroke:#01579b,stroke-width:2px

    classDef agent fill:#fff3e0,stroke:#e65100,stroke-width:2px

    classDef output fill:#e8f5e9,stroke:#2e7d32,stroke-width:2px

    classDef router fill:#f3e5f5,stroke:#6a1b9a,stroke-width:2px

    

    class BloggerCrew,LinkedInCrew,ResearcherCrew crew

    class BA1,BA2,BA3,LA1,LA2,RA1,RA2 agent

    class BlogOutput,LinkedInOutput,ResearchOutput output

    class Router router

```



## 🚀 Features



- **Multi-Crew Architecture**: Three specialized crews for different content types

- **Flow-Based Orchestration**: Dynamic routing based on user selection

- **High-Quality Content Generation**: Leverages Gemini 2.5 Flash with 100K token capacity

- **Professional Standards**: Blog crew simulates NYT editorial process

- **Web Research Integration**: Uses SerperDevTool for current information

- **Persistent State Management**: Tracks all generated content across the flow



## 📋 Content Types



### 1. Blog Post Generation

- **Crew**: Blogger Crew (3 agents)

- **Output**: 1,200-1,500 word NYT-style technology analysis

- **Process**: Research → Writing → Editorial Review

- **Agents**:

  - Senior Investigative Technology Journalist

  - Senior Staff Writer - Technology Section  

  - Editorial Director & Senior Editor



### 2. LinkedIn Post Generation

- **Crew**: LinkedIn Post Generator Crew (2 agents)

- **Output**: Engaging professional social media content

- **Process**: Research → Content Creation

- **Agents**:

  - Senior LinkedIn Post Researcher

  - Senior LinkedIn Post Writer



### 3. Research Report Generation

- **Crew**: Researcher Crew (2 agents)

- **Output**: Executive summary with key insights

- **Process**: Deep Research → Report Generation

- **Agents**:

  - Senior Content Researcher

  - Senior Content Report Generator



## Installation



Ensure you have Python >=3.10 <3.14 installed on your system. This project uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.



First, if you haven't already, install uv:



```bash

pip install uv

```



Next, navigate to your project directory and install the dependencies:



(Optional) Lock the dependencies and install them by using the CLI command:

```bash

crewai install

```



### Configuration



**1. Create and add your API keys into the `.env` file**



```bash

# Google AI API key (required)

GOOGLE_API_KEY=your_google_api_key_here



# Serper API key for web search (required)  

SERPER_API_KEY=your_serper_api_key_here

```



- Get your Google AI API key from: https://aistudio.google.com/app/apikey

- Get your Serper API key from: https://serper.dev/



**2. Project Structure**



```

content_creator_flow/

├── src/content_creator_flow/

│   ├── main.py              # Main flow orchestration

│   ├── crews/               # Individual crew configurations

│   │   ├── blogger/         # Blog post generation crew

│   │   ├── linkedin_post_generator/  # LinkedIn content crew

│   │   └── researcher/      # Research report crew

│   └── tools/               # Shared tools and utilities

├── output/                  # Generated content output

│   ├── blog_post.md

│   ├── linkedin_post.md

│   └── report.md

└── crewai_flow.html        # Flow visualization



## Running the Project



### Running the Flow



From the root folder of your project:



```bash

crewai run

```



Or using the alternative command:



```bash

python -m content_creator_flow.main

```



### Usage Example



When you run the project, you'll be prompted with:



```

🚀 CONTENT CREATOR FLOW

==================================================



📝 STEP 1: Topic Selection

-------------------------

Enter a topic for content creation: AI and Ethics



📋 STEP 2: Content Type Selection

--------------------------------

   1. blog post

   2. linkedin post

   3. research report



Enter your choice (1, 2, or 3): 1



🔄 Starting AI workflow...

```



The system will then:

1. Route your request to the appropriate crew

2. Execute the multi-agent workflow

3. Save the generated content to the `output/` directory



### Visualizing the Flow



To generate a visual representation of the flow:



```bash

crewai plot

```



This creates `crewai_flow.html` which you can open in a browser to see the interactive flow diagram.



## 🔧 Advanced Configuration



### Customizing Agents



Each crew has its own configuration files:

- `src/content_creator_flow/crews/[crew_name]/src/[crew_name]/config/agents.yaml`

- `src/content_creator_flow/crews/[crew_name]/src/[crew_name]/config/tasks.yaml`



### Flow State Management



The flow uses Pydantic models for state management (main.py:15-21):

```python

class ContentCreatorState(BaseModel):

    researcher_report: str = ""

    linkedin_post: str = ""

    blog_post: str = ""

    content_type: str = ""

    topic: str = ""

```



### Adding New Content Types



1. Create a new crew directory under `src/content_creator_flow/crews/`

2. Define agents and tasks in the config files

3. Add a new route in `main.py` router function

4. Create a listener method for the new crew



## 📊 Example Outputs



### Research Report

- Executive summary format

- Key insights and analysis

- Structured with title, summary, insights, conclusion, and sources



### Blog Post  

- 1,200-1,500 words

- NYT Technology section style

- Includes byline, compelling headline, and expert quotes



### LinkedIn Post

- Concise and engaging

- Professional tone

- Includes relevant hashtags

- Call-to-action for engagement



## 🛠️ Troubleshooting



### Common Issues



1. **API Key Errors**: Ensure your `.env` file contains valid API keys

2. **Import Errors**: Run `crewai install` to ensure all dependencies are installed

3. **Token Limits**: The Gemini model has a 100K token limit - very long topics may need to be broken down



### Debug Mode



For verbose output during execution, agents and tasks are configured with `verbose=True` by default.



## 📈 Performance Considerations



- **LLM**: Uses Gemini 2.5 Flash for fast, cost-effective generation

- **Token Usage**: Each crew is configured with appropriate token limits

- **Sequential Processing**: Agents work sequentially to ensure quality control

- **Output Caching**: Generated content is saved to disk for reuse



## 🤝 Contributing



Feel free to submit issues, fork the repository, and create pull requests for any improvements.



## 📚 Resources



- [crewAI Documentation](https://docs.crewai.com)

- [crewAI GitHub](https://github.com/joaomdmoura/crewai)

- [Google AI Studio](https://aistudio.google.com/app/apikey)

- [Serper API](https://serper.dev/)



## 📄 License



This project is licensed under the MIT License - see the LICENSE file for details.



---



Built with ❤️ using crewAI - Let's create amazing content together!