agent evaluation eval risk: low
Comprehensive Final Review Checklist Assessor
Perform a comprehensive final review merging all work streams using a checklist to confirm technical feasibility, creative vision alignment, requirements met, quality standards ach…
PROMPT
Perform a comprehensive final review merging all work streams. Review checklist: - Technical feasibility confirmed - Creative vision aligned - All requirements met - Quality standards achieved - Consistency across all elements - Ready for publication Provide a final assessment with any last recommendations.
REQUIRED CONTEXT
- all work streams
EXPECTED OUTPUT
- Format
- structured_report
- Constraints
-
- address each checklist item
- include final assessment
- include recommendations
SUCCESS CRITERIA
- Confirm technical feasibility
- Align creative vision
- Meet all requirements
- Achieve quality standards
- Ensure consistency across all elements
- Verify ready for publication
FAILURE MODES
- Assumes undefined prior work streams
- Vague checklist without detailed criteria
- May overlook unmentioned issues
CAVEATS
- Dependencies
-
- Requires prior context on all work streams
- Missing context
-
- Project background and description of work streams.
- Specific requirements, creative vision, and quality standards.
- Target output format (e.g., structured report).
- Ambiguities
-
- "Work streams" not defined or described.
- Checklist items like "Technical feasibility", "Creative vision", "requirements", "Quality standards", "elements" lack definitions or context.
QUALITY
- OVERALL
- 0.55
- CLARITY
- 0.85
- SPECIFICITY
- 0.50
- REUSABILITY
- 0.40
- COMPLETENESS
- 0.45
IMPROVEMENT SUGGESTIONS
- Add a placeholder like '[INSERT PROJECT CONTEXT AND WORK STREAMS HERE]' at the beginning.
- Specify output format: 'For each checklist item, state Yes/No with explanation, then overall assessment and recommendations.'
- Include examples of a final assessment for better guidance.
USAGE
Copy the prompt above and paste it into your AI of choice — Claude, ChatGPT, Gemini, or anywhere else you're working. Replace any placeholder sections with your own context, then ask for the output.
MORE FOR AGENT
- Advanced LLM-as-Judge Evaluation agent evaluation
- Context Degradation Pattern Diagnoser agent evaluation
- Agent Evaluation Framework Builder agent evaluation
- LLM Experiment Integrity Auditor agent evaluation
- Experiment Results Claim Evaluation Gate agent evaluation
- NVIDIA Cosmos Policy LIBERO RoboCasa Evaluator agent evaluation
- Nielsen Heuristics Mobile UX Auditor agent evaluation
- Scientific Manuscript Peer Reviewer agent evaluation
- Nielsen Heuristics Mobile UX Auditor agent evaluation
- UX/UI Principles Agent Skills agent evaluation
- Nielsen Heuristics Mobile UX Auditor agent evaluation
- ScholarEval Academic Research Evaluator agent evaluation
- AI Agent Reasoning Trace Optimizer agent evaluation
- Experiment Integrity Cross-Model Auditor agent evaluation
- Research Result-to-Claim Evaluator agent evaluation
- Web Interface Guidelines File Reviewer agent evaluation
- UX/UI Principles Interface Evaluator agent evaluation
- UX/UI Principles Agent Skills agent evaluation
- Conductor Project Artifacts Validator agent evaluation
- Comprehensive Codebase Bug Analysis and Fixer agent analysis
- Xcode MCP Usage Guidelines for Agents agent tool_use
- Xcode MCP Usage Guidelines agent tool_use
- Rapid App MVP Prototyper agent coding
- Local Documentation Online Sync Automator agent operations
- HashiCorp Packer Golden Image Expert agent operations