forge-benchmark

Validation posture scoring analyzes e2e-evidence, validation plans, hook definitions, and run metadata to compute weighted grades across coverage, evidence quality, enforcement, and speed dimensions. This generates a dated JSON benchmark report.

Validation posture benchmarking objectively measures your project's testing health by quantifying coverage, evidence quality, enforcement, and execution speed. You need it to track validation trends over time or identify gaps against baseline targets.

Run the benchmark scoring script after a validation run to scan directories, parse .vf/last-run.json for duration, and compute weighted dimension scores. It outputs a dated JSON artifact with actionable recommendations for release notes.

The scoring script checks hook and rule files alongside e2e-evidence and validation plans to measure enforcement. Projects should maintain these definitions and generate .vf run metadata to receive accurate enforcement and speed scores.

Benchmark scores apply specific weights: Coverage at 35%, Evidence Quality at 30%, Enforcement at 25%, and Speed at 10%. These weighted dimensions aggregate into a final letter grade from A to F.

Yes, running the scoring script in CI after validation runs produces dated JSON benchmark artifacts. Teams track these reports over time to monitor coverage trends, compare against baselines, and attach them to release notes.