skill-validate

Run multi-AI validation by invoking the validate command on your target file or directory, which coordinates multiple AI models to evaluate code quality, security, best practices, and completeness while generating a unified report with a 75/100 pass threshold.

Yes, automated code validation supports targeted focus areas such as security and performance. You can apply the validation workflow to specific files or directories and optionally compare results against a reference baseline.

Debate-driven quality scoring works by coordinating multiple AI models to evaluate and debate code across four dimensions. This multi-AI debate extracts objective issues and generates actionable recommendations within a detailed validation report.

The pass threshold for multi-AI code validation is 75 out of 100. The validation process scores code quality across four dimensions and generates a detailed report indicating whether this threshold is met.

Multi-AI validation works with both individual files and entire directories. It supports comprehensive validation of code repositories and components, allowing you to target specific files or validate broader directory structures.

The best way to compare code quality against a reference baseline is to use validation workflows that support reference comparisons. This multi-AI approach debates and scores code across four dimensions, extracting issues and providing objective quality assessment against your baseline.