| name | design-grill |
| description | Use when a design needs pressure-testing for UX, information architecture, states, and accessibility. |
design-grill
Lifecycle stage: CHALLENGE
Trigger
Use before implementing user-facing flows.
When not to use
Do not use when this trigger is absent; choose the command or skill that owns the requested state, artifact, and verification gate.
Inputs
- User request or current artifact.
- Known constraints and context.
- Relevant evidence or source links, if available.
- Current Agent Brain state.
Procedure
- State the current state and target artifact.
- Identify missing blockers and ask at most three blocking questions.
- Separate facts, assumptions, hypotheses, and open questions.
- Apply the anti-rationalization table below.
- Produce the required artifact: Design Grill Report.
- Add evidence, risks, decision, and next state.
Anti-Rationalization
| Shortcut | Rebuttal |
|---|
| "This is obvious." | Write the assumption and evidence. If you cannot, it is not obvious. |
| "We can do this later." | If the missing step changes the decision, do it now or state the risk. |
| "The user wants speed." | Reduce scope; do not skip the quality bar. |
| "This does not need verification." | Every important claim or behavior needs proof. |
Verification
- Required artifact exists and is named.
- Facts, assumptions, and open questions are separated.
- Evidence or evidence gaps are explicit.
- Next state is stated.
- Stop conditions are honored.
Output Artifact
Design Grill Report
Use templates/grill-report.md for a concise, auditable artifact with evidence, blockers, and next action so another agent can resume.
Failure Modes
- Producing advice instead of an artifact.
- Accepting user assumptions without challenge.
- Skipping evidence because the task feels simple.
- Recommending an agent when a simpler system is enough.
- Hiding risks or open questions.
Example
Trigger: workflow or interface design has unclear states, users, or failure paths. Action: pressure-test UX, information architecture, accessibility, edge states, and safer non-agent alternatives. Output artifact: templates/grill-report.md with blockers and next action. Verification: cite flows reviewed, rejected assumptions, accessibility checks, and remaining risks.
Input: "Build an assistant that handles customer refunds."
Better response pattern:
- identify this as high-risk automation,
- ask who approves refunds and what policies apply,
- consider non-agent alternatives like form plus approval queue,
- require human-in-loop gate,
- produce the relevant artifact and next state instead of building immediately.
