Draft, Critique, and Revise Chain

Overview

Single pasteable recipe that drafts, critiques, then revises against criteria.

Use cases

Prompt

Copied 0

Do this in three clearly labeled stages:

STAGE 1 — DRAFT the deliverable for TASK.
STAGE 2 — CRITIQUE the draft against CRITERIA (bullets; be harsh).
STAGE 3 — REVISE the draft fixing every critique. Output only the revised deliverable after stage 3.

TASK:
<<<>>>

CRITERIA:
<<<>>>

CONTEXT:
<<<>>>

Can AI?

Draft, critique, and revise a deliverable in one goPaste into a general LLM and single pasteable recipe that drafts, critiques, then revises against criteria

Alternatives

Anti-Hallucination Verification ProtocolRequire verification-first behavior to reduce unsupported or hallucinated answers.ai agent framework
Critique a prompt for ambiguity and conflictsThis prompt checks a prompt written for an LLM for unclear wording, missing definitions, instructions that clash, and assumptions left unstated. It returns a list of issues, suggested edits to specific lines, and a rewritten version.ai agent framework
Few-Shot Classifier TemplateClassify items into labeled categories using few-shot examples and confidence scores.ai agent framework
Grade an AI response against expert workThis prompt rates an AI answer on how well it follows its original instructions, measured against what a top human expert would deliver. It returns a letter grade from A+ to F, five reasons for the grade, a score out of 100, and how the score was reached.ai agent framework
Grade an Output with a Custom RubricGrade an LLM or human output against a custom rubric.ai agent framework
Improve a Weak LLM PromptRewrite a weak prompt into a clearer, higher-signal prompt.ai agent framework