It runs fully offline (no telemetry, no accounts) and can export structured results as CSV, JSON, or PDF with integrity hashes (SHA-256).
It was originally built for repeatable QA validation and archival workflows where deterministic results matter more than headless automation.
Architecture notes (for those interested): - Electron (sandboxed, context-isolated) - No use of eval() - Deterministic scoring of candidate endpoints - Optional CLI mode for batch processing
I'm mainly looking for feedback on: - Whether this is useful outside of QA / archival contexts - Expectations around cross-platform support - One-time license vs SaaS for a desktop utility