Psykos is an automated bot that fetches random images from Tumblr, analyzes their aesthetics, and posts them on Bluesky. It uses advanced APIs and libraries to ensure an efficient and secure experience.
- Fetches random images from Tumblr blogs.
- Filters images based on aesthetics using the CLIP model.
- Detects unwanted themes in captions and images.
- Publishes images on Bluesky with clean captions and alt text.
- Logs history of approved and rejected posts.
- Python 3.8 or higher.
- Tesseract OCR installed (Installation Guide).
- Credentials for Tumblr and Bluesky APIs.
-
Clone this repository:
git clone https://github.com/seu-usuario/psykos.git cd psykos -
Install the dependencies:
pip install -r requirements.txt
-
Create and configure the
.envfile with your credentials:BLUESKY_HANDLE=your_user.bsky.social BLUESKY_APP_PASSWORD=your_app_password ([Get one here](https://bsky.app/settings/app-passwords)) TUMBLR_CONSUMER_KEY=your_consumer_key TUMBLR_SECRET_KEY=your_secret_key
-
Make sure Tesseract OCR is installed and set the path in the
config.pyfile:TESSERACT_CMD = r"C:\Program Files\Tesseract-OCR\tesseract.exe"
-
Create and add the Tumblr blog titles to the
tumblrs.txtfile, separated by commas:blog1, blog2, blog3 -
Run the bot:
python main.py
-
The bot will fetch images, apply filters, and post to Bluesky automatically.
Contributions are welcome! To contribute:
- Fork the project.
- Create a branch for your feature:
git checkout -b my-feature
- Make your changes and commit:
git commit -m "Feature description" - Push your changes:
git push origin my-feature
- Open a Pull Request.
This project is licensed under the GNU GPLv3 License.