Skip to main content

Agents

Purpose​

Use Agents to store the runnable logic that processes dataset inputs.

When to use it​

Create or edit an agent whenever you want to change the behavior being evaluated.

Prerequisites​

  • You have a project.
  • You know what inputs the agent is expected to handle.

Steps​

  1. Open Agents inside a project.
  2. Create a new agent and give it a descriptive title.
  3. Add or edit the agent content.
  4. If needed, configure any required rate limit settings.
  5. Open the agent detail page to review configuration, content, required secrets, and linked eval usage.
  6. Use the agent run page for a quick preview before attaching the agent to an eval.

Expected result​

You have a saved agent that can be previewed directly or used in formal eval runs.

What to know​

  • Agent preview is blocked when required secrets are missing.
  • Agent runs can capture output, errors, and console logs.
  • Agents can be reused across multiple evals.