infer-enriched-capabilities-from-code

Capability inference generates traceable YAML artifacts and decision manifests that link business-rule overrides and warnings to explicit scan-index sources. It validates inputs, enforces depth-cap constraints, and computes caps from code signals for deterministic artifact production.

Capability inference requires scan-index.json, upstream codify proposals, and an output path under the STM evidence directory. It applies after infer-scope-from-code during the codify play to validate resolution-protocol requirements and compute depth-spectrum caps from code signals.

Yes, codebase scanning extracts business-rule overrides by merging domain KB base values with project-specific scan evidence. The resulting enriched capability artifacts are required specifically for downstream intent-epic generation during brownfield codify bootstrapping.

Decision traceability works by producing a decision manifest that links each override and warning to explicit scan-index or KB feature sources. This ensures evidence-grounded decisions by validating inputs and emitting traceable YAML artifacts under the STM area for the product-keeper.