20-Case Agent Eval Set with Pass/Fail Rubric for Normal, Edge and Unsafe Requests
Builds 20 test cases for your agent's task, with expected behavior, failure signals and a pass/fail rubric covering normal, edge and unsafe requests.
260 prompts
Prompts written for Gemini, Google's assistant. They cover writing, research and image work; the image prompts are written for Gemini's image generation and editing, where you can upload a photo and describe the change.
2 results
Builds 20 test cases for your agent's task, with expected behavior, failure signals and a pass/fail rubric covering normal, edge and unsafe requests.
Produces a set of testable guardrail rules for your coding agent, covering blocked commands, approval gates, secret handling and hook checks.
Yes, with the image prompts that say so. Upload the photo in the Gemini app, paste the prompt, and fill in the blanks that describe what to keep and what to change.
Gemini doesn't accept a prompt in its link. The button copies the prompt first and then opens Gemini, so you paste it into the message box.
Yes. They are plain text, so you can paste them into AI Studio or send them as the user message through the Gemini API. Image prompts need a model that can generate images.