multi-judge-mission-review

What problem does it solve?

Mission-level evaluation of research or product missions by running three independent judge personas to produce a unified score and section-anchored critique for leadership sign-off.

Core Features & Use Cases

• Tri-judge evaluation: three personas (self, external, critical) independently score and critique the same mission report.
• Structured output: returns a JSON object with fields scope, judgeId, score, critique, organized by report sections.
• Governance-ready feedback: supports leader aggregation of scores and justified recommendations for decision-making.
• Clear inputs & rules: accepts inputs such as topic, report, styleProfile, lengthProfile, and depth; enforces the outlined hard rules for critique generation.

Quick Start

Submit the mission report and select a judge persona to generate a JSON score and critique.