It's Important!
If you've built applications or AI agents, even using no-code or low-code tools, that use institutional data, connect to Lehigh systems, or will be available to others at Lehigh, Information Technology Services (ITS) is here to help! To ensure that your agent remains reliable, secure, portable to websites, and available for the long term, we follow a simple review and deployment workflow together.
Key Requirements for Deployment
- Shared Code & Documentation: Your application/agent's code or configuration must be saved in Lehigh's central project workspace (GitLab). You'll need to include a short overview explaining what the tool does, how it was created, data uses, and what AI models or systems/tools it connects with.
- Security Review: ITS will programmatically perform security reviews.
- Standardized Look & Feel: If your agent appears on a website, ITS will help ensure it matches Lehigh's branding and web standards.
- Human in the Loop: Agents can look up information and make recommendations, but they are not allowed to make automatic changes to university systems. A human user must always review and perform actions that alter data.
- Policy Compliance: Agents must follow all standard Lehigh policies, including guidelines on ethical research and university data handling.
How to Get Started
Don't worry if you aren't familiar with version control tools like Git or GitLab. The process starts with a simple intake meeting where ITS staff will help you set up a repository, document your agent, and guide you step-by-step through the deployment process.