linkAIin - LinkedIn AI Auto-Poster

A Python-powered Google Cloud Function that automatically generates and posts engaging content to LinkedIn using AI. This project leverages OpenAI's GPT models to create professional, data-driven LinkedIn posts based on specified topics, and can optionally include AI-generated images.

Features

• 🤖 AI-powered LinkedIn content generation
• 🔍 Web research capability to incorporate relevant information
• 🖼️ Optional AI image generation using OpenAI's image models
• 📊 Engagement analysis for created content
• 📧 Email delivery option as an alternative to direct LinkedIn posting
• 🔗 Integration with external web content through provided links
• ✏️ Customizable post length

How It Works

1. Research Phase: The system searches the web for content related to your topic, with special focus on any links you provide.
2. Content Generation: Using OpenAI's models, it creates an engaging LinkedIn post based on the research.
3. Image Creation: Optionally generates a relevant image for your post.
4. Engagement Analysis: Analyzes the potential engagement of your post.
5. Delivery: Either posts directly to LinkedIn OR sends the content via email.

Prerequisites

• Google Cloud Platform account
• LinkedIn Developer account with API access (for posting to LinkedIn)
• OpenAI API key
• Python 3.9+
• For email delivery: App password for your email account

Local Setup

1. Clone this repository:

git clone https://github.com/federicocesarini1/linkAIin.git
cd linkAIin

2. Create a virtual environment (recommended):

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file for local environment variables:

OPENAI_API_KEY=your_openai_api_key

5. Run the function locally using Functions Framework:

functions-framework --target=linkedin_ai_poster --debug

6. Test the function with curl (or use Postman):

curl -X POST http://localhost:8080 \
  -H "Content-Type: application/json" \
  -d '{
    "openai_api_key": "your_openai_api_key",
    "topic": "Latest trends in AI engineering",
    "links": ["https://example.com/article1"],
    "generate_image": true,
    "max_length": 2000,
    "post_to_linkedin": true,
    "linkedin_token": "your_linkedin_token"
  }'

Email Delivery Option

If you prefer to receive the generated content via email instead of posting directly to LinkedIn:

curl -X POST http://localhost:8080 \
  -H "Content-Type: application/json" \
  -d '{
    "openai_api_key": "your_openai_api_key",
    "topic": "Latest trends in AI engineering",
    "links": ["https://example.com/article1"],
    "generate_image": true,
    "max_length": 2000,
    "post_to_linkedin": false,
    "send_email": true,
    "email_app_password": "your_email_app_password",
    "destination_email": "your_email@example.com"
  }'

Deploying to Google Cloud Functions

1. Ensure you have the Google Cloud SDK installed and configured.

2. Deploy the function:

gcloud functions deploy linkedin_ai_poster \
  --runtime python39 \
  --trigger-http \
  --allow-unauthenticated

3. For better security in production, add authentication:

gcloud functions deploy linkedin_ai_poster \
  --runtime python39 \
  --trigger-http \
  --no-allow-unauthenticated

4. Set necessary environment variables (optional but recommended):

gcloud functions deploy linkedin_ai_poster \
  --update-env-vars OPENAI_API_KEY=your_openai_api_key

API Parameters

The function accepts the following parameters via POST request:

Parameter	Type	Required	Description
openai_api_key	string	Yes	OpenAI API key for AI content generation
topic	string	Yes	Main topic/subject for the post
links	array	No	List of URLs to include in research
generate_image	boolean	No	Whether to generate an AI image (default: false)
max_length	integer	No	Maximum character length for the post (default: 3000)
post_to_linkedin	boolean	No	Post content to LinkedIn (default: true)
linkedin_token	string	Conditional	Required if post_to_linkedin is true
send_email	boolean	No	Send content via email (default: false)
email_app_password	string	Conditional	Required if send_email is true
destination_email	string	Conditional	Required if send_email is true

Getting a LinkedIn Access Token

To get a LinkedIn access token for posting:

1. Create a LinkedIn Developer application at https://www.linkedin.com/developers/
2. Set up OAuth 2.0 settings with appropriate scopes (r_liteprofile, w_member_social)
3. Use the authorization flow to get your access token:

curl -X POST https://www.linkedin.com/oauth/v2/accessToken \
  -H 'Content-Type: application/x-www-form-urlencoded' \
  -d 'grant_type=authorization_code&code=YOUR_AUTH_CODE&client_id=YOUR_CLIENT_ID&client_secret=YOUR_CLIENT_SECRET&redirect_uri=YOUR_REDIRECT_URI'

Security Considerations

• Never commit API keys or tokens to the repository
• Use environment variables or secure secret management for sensitive data
• In production, implement proper authentication for the function endpoint
• Consider adding rate limiting to prevent abuse

Project Structure

• main.py: Main Cloud Function entry point
• config.py: Configuration settings
• ai_agents.py: Definitions for AI agents handling various tasks
• linkedin_api.py: LinkedIn API integration

Limitations

• LinkedIn access tokens expire after a limited time and need to be refreshed
• OpenAI API usage incurs costs based on token consumption
• LinkedIn API limitations may apply based on your developer account

License

MIT License

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.