When the wizard asks clarifying questions
2 min
The /scaffold.clarify probe checks if your description / PRD has enough signal. Answer the questions to lock the plan; skip to fall through.
Avery's wizard runs a `/scaffold.clarify` probe before committing to a plan. The local SLM reads your input + flags ambiguous spots: missing entity types, undefined enums, unclear roles, vague non-functional requirements.
If the probe says `sufficient: false`, the wizard pops a Clarifying Questions modal listing each question with a free-text answer field. Three options:
1. Answer every question. The planner re-runs with the answers prepended to your input. This is the highest-quality path — clarifying answers carry as much weight as the original PRD.
2. Answer some + skip the rest. The planner runs with what you gave; skipped questions become assumed defaults (the planner picks something reasonable). Less precise than option 1 but faster.
3. Skip all + 'Use my input as-is'. The planner runs with no clarification. Faster still; quality matches what your raw input deserves.
What triggers clarifying questions?
• Entities mentioned without fields. • Roles mentioned without permission boundaries. • External services mentioned without a specific provider ("send email" but no Resend / Gmail / SMTP pin). • Non-functional requirements that need parameters ("daily reports" — at what time? to whom?).
If you're getting a lot of clarifying questions, that's a signal your PRD/description is missing concrete specs. Add a paragraph for the affected entity / role / integration and re-run.
Live recipes need the desktop
This article is a static preview. The in-app Help sidecar inside Avery NXR can fire each step against your live project — install the desktop to use it interactively.