Agent Test Case Suite Designer

Build a set of test cases that cover normal use, edge cases and failures.

Added October 15, 2026. Written with AI assistance and checked for clarity and placeholders; not reviewed by a lawyer, accountant, clinician or other professional. How these prompts are made

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Design a test suite for an agent that [AGENT_PURPOSE]. Tools it uses: [TOOLS]. Users and their typical inputs: [USERS_AND_INPUTS]. Failures that would hurt most: [COSTLY_FAILURES].

Create 30 test cases grouped into: normal tasks (8), ambiguous requests (5), missing information (4), conflicting instructions (3), tool failures (4), policy boundaries (4), and long or messy inputs (2). For each give an id, the input, the expected behaviour in one or two sentences, what counts as pass or fail, and the severity if it fails. Include checks for the agent's honesty about what it did and did not do. Add a scoring sheet with columns for run date, version, result and notes, and a rule for what pass rate is required before release. Suggest how to expand the suite from real failures, and how to keep private data out of test cases.

Return the cases as a table, then a short plan for running the suite weekly.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.