meow:elicit

Structured second-pass reasoning re-examines existing outputs through named reasoning methods like pre-mortem or inversion to surface deeper insights. It applies a deterministic workflow to load context, select a method, and produce lens-based findings.

To validate a plan, provide a selected reasoning lens on the current output to generate structured findings. The workflow guides you through loading context, selecting a method like first principles, and applying the lens.

You need adversarial testing when deeper analysis is required after a review verdict, plan creation, or any agent output. It uses reasoning methods like red team and socratic lenses to surface deeper insights and validate outputs.

Yes, you can append analyses without altering the original verdict by integrating with the review cycle. This structured elicitation workflow loads existing context and produces separate lens-based findings for deeper analysis.

Available reasoning methods for lens-based analysis include pre-mortem, inversion, red team, socratic, and first principles. These named lenses guide the structured elicitation workflow to re-examine existing outputs and surface deeper insights.

No external dependencies or components are required for structured elicitation. The workflow operates independently to load context, select a reasoning method, apply a lens, and produce structured findings from existing outputs.