v0.3.0

Internal refactor with zero behaviour change: rule metadata now comes from YAML.

Highlights

• Rule metadata from YAML — rules/pfmea_control_plan_rules.yaml is the single source of truth for each rule's id, severity, title, message template, description and rationale. The checker reads it instead of hardcoding these values. (#1)
• Validation behaviour remains unchanged — same finding types, severities, score and verdict; Markdown and JSON output identical.
• Evaluation logic remains in Python — this is intentionally not a generic rule engine.
• Parity tests protect existing outputs — the seeded example, a clean case and a warnings case are locked against the v0.2 behaviour.

No change to scoring, matching, finding types, the JSON schema or the CLI.

⚠️ Surfaces potential findings to support human review — it does not replace human technical judgement and makes no regulatory/normative conformance claim.

See the CHANGELOG.

v0.2.0

Interoperability and real-world workbook compatibility, kept focused on PFMEA ↔ Control Plan validation.

Highlights

• JSON report output for automation-friendly reports via --format json (Markdown stays the default). Includes metadata, inputs, verdict, score, a summary (by severity and by finding type) and the full findings list. (#3)
• Broader column alias support for real-world PFMEA and Control Plan headers — e.g. Operation No/Op #/Process No, Severity Rating, Detection Method, Special Characteristics/Key Characteristic/Critical Characteristic, Control Technique/Measurement Technique, Out of Control Action. (#2)
• Explicit sheet selection for multi-sheet workbooks via --pfmea-sheet / --control-plan-sheet; a missing sheet lists the available ones. (#4)

No change to scoring, matching, finding types or the JSON schema.

⚠️ Surfaces potential findings to support human review — it does not replace human technical judgement and makes no regulatory/normative conformance claim.

See the CHANGELOG.

v0.1.0 — PFMEA ↔ Control Plan checker

First public release of quality-docs-validator — a local-first tool that detects potential inconsistencies between a PFMEA and a Control Plan before audits, customer submissions or production issues.

Highlights

• Reads PFMEA + Control Plan .xlsx (recommended template + column aliases) and matches rows by operation.
• Six explicit, documented checks: UNMATCHED_PROCESS_STEP, MISSING_CONTROL, SPECIAL_CHARACTERISTIC_NOT_CONTROLLED, MISSING_REACTION_PLAN, plus WEAK_DETECTION_METHOD and HIGH_SEVERITY_WEAK_CONTROL as warnings.
• Severity-weighted score (0–100) with PASS / PASS-WITH-WARNINGS / NEEDS-REVIEW / FAIL bands.
• Markdown report + rich terminal summary; bundled synthetic examples with a seeded gap.
• CI green on Windows + Linux, Python 3.12 + 3.13.

Quickstart

pip install -e .
qdv pfmea-control-plan --pfmea examples/pfmea.xlsx --control-plan examples/control-plan.xlsx --out report.md

⚠️ Surfaces potential findings to support human review — it does not replace human technical judgement and makes no regulatory/normative conformance claim.

See CHANGELOG.md and docs/FINDINGS.md.