Self-Critique Verification Gate For Agent Output
Adds a rigorous self-review pass where the agent critiques its own draft action against a checklist before committing it.
Prompt
ROLE: You are a reliability engineer who designs verification gates that catch agent mistakes before they execute. CONTEXT: My agent produces [OUTPUT_TYPE] that, if wrong, causes [CONSEQUENCE]. I want a mandatory self-critique step before the output is finalized or any irreversible action runs. TASK: Design and run the verification gate on the draft I will provide as [DRAFT]. 1. First, restate the task's true success criteria in your own words. 2. Critique the draft against this checklist: correctness, completeness, adherence to constraints, safety/side-effects, and assumptions made without evidence. 3. For each issue found, rate severity (blocker/major/minor) and propose a concrete fix. 4. Decide a verdict: APPROVE, REVISE, or ABORT (with reason). 5. If REVISE, produce the corrected output. OUTPUT FORMAT: A structured report: Success Criteria | Findings table (Issue | Severity | Fix) | Verdict | Revised Output (if any). CONSTRAINTS: Be adversarial toward the draft, not charitable. Do not approve anything with an unresolved blocker. If evidence is missing to judge correctness, mark it as an explicit assumption rather than guessing. DRAFT: [DRAFT]
How to use this prompt
- 1
Copy the prompt above and paste it into ChatGPT, Claude, or Gemini — or open it in the visual Studio to edit each part on a canvas and run it with your own key.
- 2
Replace any bracketed placeholders with your specifics. The more concrete your context and constraints, the sharper the result — see the 5-part prompt structure.
- 3
Run it, then refine. Ask the model to critique and improve its own answer with self-critique prompting.
Techniques in this prompt
Has the model critique its own draft against criteria, then revise — raising quality in a single pass.
Learn this techniquePins the response to a defined structure so it drops straight into your workflow.
Learn this techniqueForces explicit intermediate reasoning instead of jumping to a conclusion, which improves accuracy on hard tasks.
Learn this techniqueRecommended models
Build on this prompt
Open it in the visual Studio to wire it into a full workflow with your own API key — or learn the craft behind prompts like this.
More in AI Agents & Autonomous Workflows
Autonomous Agent System Prompt Architect
Designs a complete, production-grade system prompt for an autonomous agent including persona, tool contracts, guardrails, and stop conditions.
ReAct Loop Reasoning Trace Designer
Builds a strict ReAct-style Thought/Action/Observation loop with explicit formatting and self-correction rules for tool-using agents.
Multi-Agent Orchestration Blueprint
Plans a coordinated multi-agent team with roles, hand-off contracts, shared memory, and conflict resolution for a complex objective.
Agent Tool Definition And Schema Writer
Writes precise tool/function definitions with JSON schemas, descriptions, and usage hints that minimize wrong-tool and bad-argument errors.