Build a repeatable workflow

Use this topic-specific sequence: decompose → gather → calculate → answer → verify. Give each stage one input, one output, and one gate. The first run should be narrow and reversible. Later automation is earned by measured reliability, not by how easy it is to connect tools.

For make a multi-constraint laptop recommendation auditable without requesting hidden reasoning, begin with the job card and sanitized packet. Run the constrained prompt:

Compare three laptops for college using price, measured battery life, weight, and required-software compatibility. First output a short comparison plan and evidence table. Then recommend one in under 150 words with three deciding facts. Finally list unknowns to verify on seller pages. Give concise rationale, not private hidden reasoning.

Save the response beside its prompt and input version. Then apply the quality rubric and open authoritative product pages, recalculate weighted scores, check units and model variants, test software requirements, and see whether the recommendation changes under reasonable weights. A failed check returns to the smallest responsible stage; do not regenerate everything. If the source was missing, repair context. If the instruction was ambiguous, repair the prompt. If the candidate violates policy, stop and escalate rather than prompt around the policy.

Name an owner for source approval, generation, verification, and release. One person may hold several roles on a small project, but the role changes should remain visible. The reviewer needs the evidence packet, not merely the final artifact.

Define operational states: draft, needs evidence, blocked, approved, released, and rolled back. This vocabulary prevents a plausible draft from being mistaken for an approved result. Attach timeouts, retry limits, and an off switch to any automated stage.

Log the defect category rather than just “bad output.” This chapter's recurring defects are equating longer rationale with truth; post-hoc justification; invented specifications; hidden weighting; arithmetic that cannot be reproduced. Track their rate on representative cases. Review false positives and false negatives separately when classification is involved; track factual, continuity, or rights defects when producing media.

The end product is a reasoning audit card containing decomposition, evidence, concise rationale, calculations, assumptions, and independent checks. Ask for visible work products—plans, formulas, evidence tables, assumptions, and checks. A model's private reasoning is neither required nor a substitute for proof. Periodically rerun a stable set of cases after changing models, prompts, source material, formulas, or settings.

Imagine the independent check fails after release. Identify how to stop distribution, identify affected outputs, restore the last approved version, notify the owner, and preserve enough evidence to learn. A workflow without rollback is only a happy-path demo.