What problem does it solve? Teams often ship features without verifying that non-functional requirements like performance, security, reliability, and maintainability are actually met. This Skill audits existing implementation evidence against defined thresholds and produces deterministic PASS, CONCERNS, or FAIL classifications before release. ## Core Features & Use Cases - Evidence-Based NFR Audit: Evaluates test results, metrics, logs, scans, and CI data against thresholds sourced from tech specs, PRDs, and test-design documents, never guessing undefined thresholds. - Four-Domain Assessment: Runs dedicated Security, Performance, Reliability, and Maintainability evaluations with support for parallel subagent or sequential execution modes. - Release Gate Reporting: Generates a structured nfr-assessment.md report with quick wins, prioritized remediation actions, evidence gaps, and a gate YAML snippet for CI integration. - Use Case: Before releasing a bulk export feature, run the audit to discover that a PDF-rendering vulnerability and a missing circuit breaker are release blockers, with specific owners and effort estimates for each fix. ## Quick Start Ask the assistant to audit NFR evidence for the current feature using the bmad-testarch-nfr workflow and point it at your test results, metrics, and logs directories.
