code-scorecard

To generate a code quality scorecard for a polyglot repository, audit the codebase across nine dimensions and present per-ecosystem results with a unified suite summary. Score detected ecosystems like dotnet and javascript-typescript separately using deterministic JSON evidence.

An automated codebase audit evaluates nine dimensions: architecture, code quality, testing, security, error handling, documentation, dependency management, performance, and maintainability. Each dimension receives a 0–10 score backed by deterministic JSON evidence or qualitative assessment.

Deterministic code scoring uses CodeMetrics.AI outputs as authoritative evidence when available. If deterministic evidence is missing or incompatible, it falls back to qualitative scoring against published anchors and uses ecosystem-specific thresholds to derive final scores.

Yes, you can use code scorecard metrics for vendor due-diligence and security audits. The scorecard reduces guesswork by auditing a codebase across nine quality dimensions and producing clear, actionable scores backed by deterministic evidence.

The best way to run a pre-release health check on a long-lived codebase is to run an audit across nine quality dimensions, generating a 0–10 score per dimension. This produces an actionable scorecard with deterministic JSON evidence for ongoing health monitoring.