An extension for analyzing and improving AI prompt files. Works with .prompt.md, .agent.md, SKILL.md, and .instructions.md files — providing LLM-powered semantic analysis directly in VS Code.
Features
LLM-Powered Analysis (via GitHub Copilot)
Contradiction Detection — Finds logical, behavioral, and format conflicts
Semantic Ambiguity — Ambiguity analysis with rewrite suggestions
Persona Consistency — Detects conflicting personality traits and tone drift
Cognitive Load Assessment — Warns about overly complex prompts with too many nested conditions
Semantic Coverage — Identifies gaps in intent handling and missing error paths
Composition Conflict Analysis — Detects conflicts between a prompt and other prompt files it imports via markdown links
Waza Integration
Create Eval Scaffold — Generates eval files for a skill via waza new eval <skill-name>
Run Evaluation — Executes skill evaluation via waza run <eval.yaml> --context-dir <skill-dir>
Automatic Local Fallback — If waza is not on PATH, commands attempt a local fallback via go run ./cmd/waza when a sibling waza repo is available
Editor Integration
Editor Title Bar — Analyze Prompt button appears when editing prompt files
