Install any skill in seconds. Free to start, no credit card required.
Get Started Free →ISO 13485 Quality Management System implementation and maintenance for medical device organizations. Provides QMS design, documentation control, internal auditing, CAPA management, and certification support.
.claude/skills/borghei-quality-manager-qms-iso13485/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 86% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 232% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 525% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 537% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 105% | 0% |
ISO 13485:2016 Quality Management System implementation, maintenance, and certification support for medical device organizations. Covers the full lifecycle from gap analysis through certification, plus the FDA QMSR transition (effective Feb 2026), digital/electronic QMS, and AI-enabled device considerations.
Before building or assessing the QMS, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the deliverable.
bash# Generate an ISO 13485 audit checklist (clause, process, or full system) python scripts/qms_audit_checklist.py --clause 7.3 python scripts/qms_audit_checklist.py --process design-control python scripts/qms_audit_checklist.py --audit-type system --output json python scripts/qms_audit_checklist.py --interactive
Load the reference that matches the task — keep this file lean and pull detail on demand:
qms_audit_checklist.py flag reference. Read when diagnosing tool output or validating QMS completeness.In Scope:
Out of Scope:
| Skill | Integration | |-------|------------| | quality-manager-qmr | QMR oversees QMS effectiveness; management review inputs include QMS process performance metrics | | capa-officer | CAPA system (Clause 8.5) is a core QMS process; CAPA effectiveness feeds into management review | | qms-audit-expert | Internal audit program (Clause 8.2.4) evaluates QMS processes; audit findings drive CAPA and improvement | | quality-documentation-manager | Document and record control (Clause 4.2) provides the documentation foundation for the entire QMS | | risk-management-specialist | ISO 14971 risk management integrates with design control (Clause 7.3) and product realization planning (Clause 7.1) | | fda-consultant-specialist | QMSR alignment requires mapping FDA-specific requirements (MDR reporting, UDI, Part 11) beyond ISO 13485 | | regulatory-affairs-head | Regulatory strategy informs QMS scope, market-specific requirements, and certification timelines |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,076 | 20,804 | +29% | 1 | 1 | 0% | 2,831 | 5,138 | +81% | 0 | 0 | — |
case-02 | fail→fail | 26,359 | 31,849 | +21% | 1 | 1 | 0% | 4,577 | 6,789 | +48% | 0 | 0 | — |
case-03 | fail→fail | 31,835 | 25,577 | -20% | 1 | 1 | 0% | 5,423 | 6,095 | +12% | 0 | 0 | — |
case-08 | pass→pass | 4,348 | 2,717 | -38% | 1 | 1 | 0% | 787 | 2,092 | +166% | 0 | 0 | — |
case-04 | fail→pass | 20,468 | 27,906 | +36% | 1 | 1 | 0% | 3,510 | 6,514 | +86% | 0 | 0 | — |
case-05 | pass→pass | 15,826 | 17,581 | +11% | 1 | 1 | 0% | 3,105 | 4,857 | +56% | 0 | 0 | — |
case-06 | pass→pass | 1,888 | 6,715 | +256% | 1 | 1 | 0% | 342 | 2,772 | +711% | 0 | 0 | — |
case-07 | pass→pass | 4,034 | 3,842 | -5% | 1 | 1 | 0% | 754 | 2,339 | +210% | 0 | 0 | — |
case-09 | pass→pass | 3,730 | 6,759 | +81% | 1 | 1 | 0% | 627 | 2,814 | +349% | 0 | 0 | — |
case-10 | pass→pass | 5,037 | 8,267 | +64% | 1 | 1 | 0% | 732 | 3,008 | +311% | 0 | 0 | — |
case-11 | pass→pass | 2,914 | 11,155 | +283% | 1 | 1 | 0% | 403 | 3,546 | +780% | 0 | 0 | — |
case-12 | pass→pass | 4,302 | 8,114 | +89% | 1 | 1 | 0% | 636 | 3,003 | +372% | 0 | 0 | — |
case-13 | pass→pass | 3,150 | 11,175 | +255% | 1 | 1 | 0% | 559 | 3,656 | +554% | 0 | 0 | — |
case-14 | pass→pass | 7,032 | 9,220 | +31% | 1 | 1 | 0% | 1,153 | 3,085 | +168% | 0 | 0 | — |
case-15 | pass→pass | 6,570 | 7,615 | +16% | 1 | 1 | 0% | 1,075 | 2,983 | +177% | 0 | 0 | — |
case-16 | fail→pass | 4,054 | 4,144 | +2% | 1 | 1 | 0% | 688 | 2,283 | +232% | 0 | 0 | — |
case-17 | fail→pass | 2,021 | 2,462 | +22% | 1 | 1 | 0% | 328 | 2,050 | +525% | 0 | 0 | — |
case-18 | pass→pass | 4,070 | 2,664 | -35% | 1 | 1 | 0% | 646 | 2,103 | +226% | 0 | 0 | — |
case-19 | fail→pass | 2,225 | 2,093 | -6% | 1 | 1 | 0% | 319 | 2,032 | +537% | 0 | 0 | — |
case-20 | fail→pass | 12,684 | 15,360 | +21% | 1 | 1 | 0% | 1,971 | 4,042 | +105% | 0 | 0 | — |
case-21 | fail→fail | 40,615 | 33,446 | -18% | 1 | 1 | 0% | 6,194 | 7,435 | +20% | 0 | 0 | — |
case-22 | fail→fail | 61,047 | 25,192 | -59% | 1 | 1 | 0% | 5,850 | 5,766 | -1% | 0 | 0 | — |
case-23 | fail→pass | 35,458 | 27,676 | -22% | 1 | 1 | 0% | 6,195 | 6,355 | +3% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.