What problem does it solve? Writing effective prompts for AI agents is error-prone: vague instructions, missing routing, and absent quality gates cause downstream failures. This Skill operationalizes research-validated prompt construction (5-element anatomy), structured negation constraints (NPT-009/NPT-013), and a weighted 7-criterion quality rubric so prompts are specific, routable, and measurable. ## Core Features & Use Cases - Interactive Prompt Builder (pe-builder): Walks users through the 5-element prompt anatomy (routing, scope, data source, quality gate, output path) and produces XML-wrapped structured prompts with a self-review score. - NPT Constraint Generator (pe-constraint-gen): Converts intent descriptions into NPT-009/NPT-013 formatted constraints with consequence chains and constructive alternatives, wrapped in XML for agent governance YAML or rule files. - Prompt Quality Scorer (pe-scorer): Evaluates prompts against the 7-criterion rubric (C1 Task Specificity through C7 Positive Framing), computes a weighted composite score, classifies into tiers, and detects 8 anti-patterns with fixes. - Use Case: Ask the builder to construct a research prompt for a .NET microservice, then have the scorer evaluate it and iterate until the composite score reaches 90+. ## Quick Start Ask the assistant to build a structured prompt for your task, generate NPT-013 constraints for an agent, or score an existing prompt against the quality rubric.
