DigitalOcean GenAI Platform lets you build GPU-powered AI agents with fully-managed deployment. Agents can use pre-built or custom foundation models, incorporate function and agent routes, and implement RAG pipelines with knowledge bases.
Guardrails scan an agent’s input and output for sensitive and inappropriate content and override the agent’s output when it detects the specified problematic content. For example, it can prevent an agent from sharing login credentials or credit card information.
The following are the supported guardrails you can attach to your agent:
Sensitive Data: Identifies and anonymizes sensitive data categories like credit card numbers, personal identifiable information, and locations data, and more. These categories are customizable.
Jailbreak: Protects your agent from being jailbroken by bad actors or malicious inputs.
Content Moderation: Guards against agent output that falls under the following categories: violence and hate, sexual content, guns and illegal weapons, regulated or controlled substances, suicide and self-harm, and criminal planning.
You cannot delete these supported guardrails. You can duplicate and edit them, and you can delete any copies you’ve made.