Making Your First Request

​

1. Create your account, organization, and project

Create or log in to your Gegentica account. On first sign-up you’ll be prompted to create an organization and a project — this is a one-time setup step. You can create additional projects later from the Projects page.

​

2. Set up your AI Providers

Before you can approve any use case, connect the AI providers (OpenAI, Anthropic, Azure OpenAI, etc.) your agents will call. AI Providers are configured once at the organization level under Settings > AI Providers, and the credentials are shared across all projects in the org. Add the provider, paste in your provider API key, and save. You can connect as many providers as you need.

​

3. Define an AI Use Case

Within your project, go to Governance > AI Use Cases and create one. An AI Use Case is the approval boundary for a provider + model combination — you can’t create an agent against a provider or model that hasn’t been approved here first. Set the use case name, AI Provider (from the providers you connected in step 2), Allowed Models, and a Risk Rating. Submit it for confirmation if your organization requires use case approval before it can be used.

​

4. Create an AI Agent

Go to Governance > AI Agents and create one. An AI Agent is the running instance that actually serves requests — it inherits its provider, models, and default prompt from the Approved Use Case you select. You can optionally override the inherited model or prompt at this step.

​

5. (Optional) Create a Policy Set

If you want rule-based screening on this agent’s traffic, go to Governance > Policy Sets and create one. A Policy Set is a collection of Rules with a Draft → Pending Review → Active lifecycle. Once it’s Active, attach it to your agent from the agent’s settings. You can skip this step and add a Policy Set to the agent later — it isn’t required to make your first request.

​

6. Generate an Access Key

Go to Governance > Access Keys and select Create. Unlike a raw provider API key, an Access Key is scoped to specific AI Agents and Prompts, not tied directly to a provider credential. Generation is a guided flow:

1. Key Info — name the key, optionally describe it, choose an Environment, and optionally set an expiry date.
2. AI Agent Access — scope to all agents or to the specific agent you created in step 4.
3. Prompt Access — scope to all prompts or a specific subset.
4. Review & Generate — confirm and generate.

​

7. Integrate Gegentica

Once your Access Key is wired in, requests will start showing up in your project’s Dashboard and Audit Log.

​

8. Next Steps

Now that you’re up and running with Gegentica, dive into the rest of the platform: