This project is a web-based AI agent that can perform several roles to assist with software development and business tasks.
This project is a submission for the [Name of Devpost Challenge].
- Live Demo: https://yendoukoa.google.ai
- Video Walkthrough: [Link to Video Walkthrough]
If you find this project useful, please consider sponsoring us!
Our platform connects to multiple AI services to provide a suite of tools for developers and entrepreneurs. The core of the application is a Flask-based backend that serves a user-friendly frontend. The AI functionalities are powered by Google's Vertex AI, enabling features like code generation, debugging, and content creation. The platform is designed to be easily extensible, allowing for the integration of new AI-powered tools in the future.
- Software Engineer: Generates multi-section HTML and CSS for a static website based on a structured text prompt.
- Debugger: Lints HTML and CSS code to find basic errors. Can analyze pasted code or fetch a file directly from a GitHub URL.
- Marketer: Creates promotional social media posts from a business description.
- System Analyzer: Scans a website URL for broken links and suggests search queries to find solutions.
The Software Engineer agent uses a simple, indented syntax to define the components of a website. Provide a description in the "Software-Engineer" text box, and the agent will return the HTML and CSS code in the response box below.
Example Prompt:
title: My Photography Portfolio
header: Jane Doe | Photographer
section: About Me
text: I am a professional photographer specializing in landscapes.
section: Gallery
images: 4
footer: Copyright © 2024 Jane Doe
The Debugger agent can analyze code in two ways:
- Paste Code: Paste your HTML or CSS code directly into the large text area.
- Use a GitHub URL: Paste the URL of a public file on GitHub into the smaller URL input field.
The agent will automatically fetch the code from the URL and analyze it.
Enter a full website URL (e.g., https://example.com) to scan the page for broken links.
- Python 3.x
pipfor installing Python packages
-
Clone the repository:
git clone https://github.com/GYFX35/AI-services.git cd AI-services -
Set up a virtual environment (recommended):
python -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate`
-
Install Python dependencies:
pip install -r requirements.txt
-
Create a
.envfile:- Copy the
.env.exampleto.envand fill in your API keys.
cp .env.example .env
- Copy the
-
Initialize the database:
flask init-db
-
Start the server:
flask run
-
Open your web browser: Navigate to
http://127.0.0.1:5000to access the application.
This project is configured for easy use across multiple platforms.
- Git Attributes: Consistent line endings are managed via
.gitattributes. - GitLab CI/CD: A
.gitlab-ci.ymlis provided for automated builds, testing withpytest, and linting.- Simply push to GitLab to trigger the pipeline.
- Ensure you set your environment variables (from
.env.example) in GitLab CI/CD settings.
- Hosting and Functions: Use
firebase.jsonand.firebasercfor deployment. - Firebase Studio: Compatible with Firebase tools for project management.
- Deploy command:
firebase deploy
- Containerized Deployment: A
Dockerfileanddocker-compose.ymlare included. - AWS SAM (Serverless Application Model): A
template.yamlis provided for deploying as an AWS Lambda function with API Gateway. - Deploy via SAM:
sam build sam deploy --guided
- App Engine: Deploy using
app.yaml.gcloud app deploy
- Cloud Run via Cloud Build: Use
cloudbuild.yamlfor automated container builds and deployment.gcloud builds submit --config cloudbuild.yaml
- Hackathon Ready: This repository includes all necessary configuration for quick cloning and deployment during hackathons.
- Project Story: (Add your project's inspiration and goals here for your submission).
- One-Click Setup: Use the provided Docker and Cloud configurations to get your demo live in minutes.
Developed for [Hackathon Name]