The CrewAI LinkedIn Content Team App is an open-source, template application for generating LinkedIn-ready content using a team of AI agents in a structured team workflow. The app is designed demonstrate how a simple agent implementation can help teams research, create, and refine engaging social media content, emphasizing high-quality output and collaborative teamwork.
This project is built on Replit.
- Automated Research: Uses tools to fetch and analyze the latest articles and trends about a specific subject focus.
- Content Creation: Develops LinkedIn posts tailored to a professional, tech-savvy audience.
- Editorial Review: Refines and ensures the quality and brand alignment of all content.
- Output Management: Automatically saves outputs into well-organized files and directories.
The app leverages CrewAI's agents and tasks to execute a sequential process:
-
Research Phase:
- Agent:
Senior Research Analyst
- Task: Collect, analyze, and summarize relevant articles about the specified subject focus.
- Output: A markdown report with article summaries, key quotes, and analysis.
- Agent:
-
Content Writing Phase:
- Agent:
Social Media Content Writer
- Task: Create LinkedIn posts based on research findings, aligning with a specific style and voice.
- Output: Drafted LinkedIn posts.
- Agent:
-
Editing Phase:
- Agent:
Senior Editor
- Task: Review and finalize social media posts for quality, brand alignment, and accuracy.
- Output: Finalized LinkedIn content and a report summarizing the process.
- Agent:
Before running the app, ensure you have the following:
-
API Keys:
- OpenAI API Key: Required for generating content and facilitating agent workflows.
- SerperDev API Key: Required for search and web scraping functionality.
- Add these keys as environment variables or directly configure them in the script.
-
Python Packages:
- Install the required Python packages:
pip install crewai crewai_tools
- Install the required Python packages:
-
Access to Replit (optional but recommended):
- Easily deploy and test the app in a cloud-based environment.
- On Replit:
- Fork or clone the repository directly in Replit for a ready-to-run environment.
- On Another Python IDE:
- Clone the repository to your local machine using:
git clone https://github.com/your-repo/crewai-linkedin-template.git
- Open the project in your preferred Python IDE.
- Clone the repository to your local machine using:
- Create a
.env
file in the project root or set environment variables with your API keys:OPENAI_API_KEY=your_openai_api_key SERPER_API_KEY=your_serperdev_api_key```
- Edit the agent prompts in the script to suit your needs:
- Define the topic of focus.
- Specify the style of writing for LinkedIn posts.
- Add your company context (e.g., what your company does, offerings, target audience).
Step 4: Run the App
- Execute the script:
python app.py
- The results, including research reports and finalized LinkedIn posts, will be saved in the outputs/ folder with timestamps for easy organization.
The app is highly flexible:
- Adjust agents’ roles and backstories to reflect your team structure and goals.
- Modify task descriptions and expected outputs to tailor content creation workflows.
- Integrate additional tools or APIs for enhanced functionality.
This project is licensed under the MIT License. See the LICENSE file for more details.
This template is powered by CrewAI and was built using Replit.
Start creating professional LinkedIn content effortlessly with this intuitive template! 🚀