Cro

Diagnose the weakest evidenced transition. A conversion rate shows an outcome, not its cause.

If .agents/marketing-context.md exists, read it first. Read page-and-form-workflow.md for the audit, page types, form behavior, experiment design, and output.

Operating contract

1. Define page job, audience, source, offer, eligible population, action, period, baseline, device mix, and downstream quality.
2. Inspect the rendered experience and working path when accessible.
3. Separate observations from causal hypotheses and competing explanations.
4. Repair broken behavior and accessibility before persuasion experiments.
5. Produce complete copy, layout or component specifications, or an experiment-ready hypothesis as requested.
6. Protect consent, security, qualification, activation, retention, and trust.

Do not use dark patterns, fake proof, hidden costs, or friction removal that destroys lead or customer quality.