Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Senior Regulatory Affairs Manager for HealthTech and MedTech companies. Prepares FDA 510(k), De Novo, and PMA submission packages; analyzes regulatory pathways for new medical devices; drafts responses to FDA deficiency letters and Notified Body queries; develops CE marking technical documentation under EU MDR 2017/745; coordinates multi-market approval strategies across FDA, EU, Health Canada, PMDA, and NMPA; and maintains regulatory intelligence on evolving standards. Use when users need to pl
.claude/skills/alirezarezvani-regulatory-affairs-head/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | 191% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 340% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 125% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 226% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 151% | 0% |
Regulatory strategy development, submission management, and global market access for medical device organizations.
Develop regulatory strategy aligned with business objectives and product characteristics.
| Factor | 510(k) | De Novo | PMA | |--------|--------|---------|-----| | Predicate Available | Yes | No | N/A | | Risk Level | Low-Moderate | Low-Moderate | High | | Clinical Data | Usually not required | May be required | Required | | Review Time | 90 days (MDUFA) | 150 days | 180 days | | User Fee | ~$22K (2024) | ~$135K | ~$440K | | Best For | Me-too devices | Novel low-risk | High-risk, novel |
REGULATORY STRATEGY
Product: [Name] Version: [X.X] Date: [Date]
1. PRODUCT OVERVIEW
Intended use: [One-sentence statement of intended patient population, body site, and clinical purpose]
Device classification: [Class I / II / III]
Technology: [Brief description, e.g., "AI-powered wound-imaging software, SaMD"]
2. TARGET MARKETS & TIMELINE
| Market | Pathway | Priority | Target Date |
|--------|----------------|----------|-------------|
| USA | 510(k) / PMA | 1 | Q1 20XX |
| EU | Class [X] MDR | 2 | Q2 20XX |
3. REGULATORY PATHWAY RATIONALE
FDA: [510(k) / De Novo / PMA] — Predicate: [K-number or "none"]
EU: Class [X] via [Annex IX / X / XI] — NB: [Name or TBD]
Rationale: [2–3 sentences on key factors driving pathway choice]
4. CLINICAL EVIDENCE STRATEGY
Requirements: [Summarize what each market needs, e.g., "510(k): bench + usability; EU Class IIb: PMCF study"]
Approach: [Literature review / Prospective study / Combination]
5. RISKS AND MITIGATION
| Risk | Prob | Impact | Mitigation |
|------------------------------|------|--------|-----------------------------------|
| Predicate delisted by FDA | Low | High | Identify secondary predicate now |
| NB audit backlog | Med | Med | Engage NB 6 months before target |
6. RESOURCE REQUIREMENTS
Budget: $[Amount] Personnel: [FTEs] External: [Consultants / CRO]Prepare and submit FDA regulatory applications.
| Characteristic | Predicate (K213456) | Subject Device | Same? | Notes | |----------------|---------------------|----------------|-------|-------| | Intended use | Wound measurement | Wound measurement | ✓ | Identical | | Technology | 2D camera | 2D + AI analysis | ✗ | New TC; address below | | Energy type | Non-energized | Non-energized | ✓ | | | Patient contact | No | No | ✓ | | | SE conclusion | New TC does not raise new safety/effectiveness questions; bench data demonstrates equivalent accuracy (±2mm vs ±3mm predicate) |
| Milestone | 510(k) | De Novo | PMA | |-----------|--------|---------|-----| | Pre-Sub Meeting | Day -90 | Day -90 | Day -120 | | Submission | Day 0 | Day 0 | Day 0 | | RTA Review | Day 15 | Day 15 | Day 45 | | Substantive Review | Days 15–90 | Days 15–150 | Days 45–180 | | Decision | Day 90 | Day 150 | Day 180 |
| Category | Common Issues | Prevention | |----------|---------------|------------| | Substantial Equivalence | Weak predicate comparison; no performance data | Build SE table with data column; cite recognized standards | | Performance Testing | Incomplete protocols; missing worst-case rationale | Follow FDA-recognized standards; document worst-case justification | | Biocompatibility | Missing endpoints; no ISO 10993-1 risk assessment | Complete ISO 10993-1 matrix before testing | | Software | Inadequate hazard analysis; no cybersecurity bill of materials | IEC 62304 compliance + FDA cybersecurity guidance checklist | | Labeling | Inconsistent claims vs. IFU; missing symbols standard | Cross-check label against IFU; cite ISO 15223-1 for symbols |
See: references/fda-submission-guide.md
Achieve CE marking under EU MDR 2017/745.
| GSPR Ref | Requirement | Standard / Guidance | Evidence Document | Status | |----------|-------------|---------------------|-------------------|--------| | Annex I §1 | Safe design and manufacture | ISO 14971:2019 | Risk Management File v2.1 | Complete | | Annex I §11.1 | Devices with measuring function ±accuracy | EN ISO 15223-1 | Performance Test Report PT-003 | Complete | | Annex I §17 | Cybersecurity | MDCG 2019-16 | Cybersecurity Assessment CS-001 | In progress |
| Class | Clinical Requirement | Documentation | |-------|---------------------|---------------| | I | Clinical evaluation (CE) | CE report | | IIa | CE with literature focus | CE report + PMCF plan | | IIb | CE with clinical data | CE report + PMCF + clinical study (some) | | III | CE with clinical investigation | CE report + PMCF + clinical investigation |
See: references/eu-mdr-submission-guide.md
Coordinate regulatory approvals across international markets.
| Market | Size | Complexity | Recognition | Priority | |--------|------|------------|-------------|----------| | USA | Large | High | N/A | 1 | | EU | Large | High | N/A | 1–2 | | Canada | Medium | Medium | MDSAP | 2 | | Australia | Medium | Low | EU accepted | 2 | | Japan | Large | High | Local clinical | 3 | | China | Large | Very High | Local testing | 3 | | Brazil | Medium | High | GMP inspection | 3–4 |
| Document Type | Single Source | Localization Required | |---------------|---------------|----------------------| | Technical file core | Yes | Format adaptation | | Risk management | Yes | None | | Clinical data | Yes | Bridging assessment | | QMS certificate | Yes (ISO 13485) | Market-specific audit | | Labeling | Master label | Translation, local requirements | | IFU | Master content | Translation, local symbols |
See: references/global-regulatory-pathways.md
Monitor and respond to regulatory changes affecting product portfolio.
| Source | Type | Frequency | |--------|------|-----------| | FDA Federal Register | Regulations, guidance | Daily | | FDA Device Database | 510(k), PMA, recalls | Weekly | | EU Official Journal | MDR/IVDR updates | Weekly | | MDCG Guidance | EU implementation | As published | | ISO/IEC | Standards updates | Quarterly | | Notified Body | Audit findings, trends | Per interaction |
REGULATORY CHANGE IMPACT ASSESSMENT
Change: [Description] Source: [Regulation/Guidance]
Effective Date: [Date] Assessment Date: [Date] Assessed By: [Name]
AFFECTED PRODUCTS
| Product | Impact (H/M/L) | Action Required | Due Date |
|---------|----------------|------------------------|----------|
| [Name] | [H/M/L] | [Specific action] | [Date] |
COMPLIANCE ACTIONS
1. [Action] — Owner: [Name] — Due: [Date]
2. [Action] — Owner: [Name] — Due: [Date]
RESOURCE REQUIREMENTS: Budget $[X] | Personnel [X] hrs
APPROVAL: Regulatory _____________ Date _______ / Management _____________ Date _______FDA Pathway Selection
Is predicate device available?
│
Yes─┴─No
│ │
▼ ▼
Is device Is risk level
substantially Low-Moderate?
equivalent? │
│ Yes─┴─No
Yes─┴─No │ │
│ │ ▼ ▼
▼ ▼ De Novo PMA
510(k) Consider required
De Novo
or PMAEU MDR Classification
Is the device active?
│
Yes─┴─No
│ │
▼ ▼
Is it an Does it contact
implant? the body?
│ │
Yes─┴─No Yes─┴─No
│ │ │ │
▼ ▼ ▼ ▼
III IIb Check Class I
contact (measuring/
type sterile if
and applicable)
duration| Factor | Schedule Pre-Sub | Skip Pre-Sub | |--------|------------------|--------------| | Novel Technology | ✓ | | | New Intended Use | ✓ | | | Complex Testing | ✓ | | | Uncertain Predicate | ✓ | | | Clinical Data Needed | ✓ | | | Well-established | | ✓ | | Clear Predicate | | ✓ | | Standard Testing | | ✓ |
| Situation | Escalation Level | Action | |-----------|------------------|--------| | Submission rejection | VP Regulatory | Root cause analysis, strategy revision | | Major deficiency | Director | Cross-functional response team | | Timeline at risk | Management | Resource reallocation review | | Regulatory change | VP Regulatory | Portfolio impact assessment | | Safety signal | Executive | Immediate containment and reporting |
| Tool | Purpose | Usage | |------|---------|-------| | regulatory_tracker.py | Track submission status and timelines | python regulatory_tracker.py |
Regulatory Tracker Features:
Example usage:
bash$ python regulatory_tracker.py --report status Submission Status Report — 2024-11-01 ┌──────────────────┬──────────┬────────────┬─────────────┬──────────┐ │ Product │ Market │ Type │ Target Date │ Status │ ├──────────────────┼──────────┼────────────┼─────────────┼──────────┤ │ WoundScan Pro │ USA │ 510(k) │ 2024-12-01 │ On Track │ │ WoundScan Pro │ EU │ MDR IIb │ 2025-03-01 │ At Risk │ │ CardioMonitor X1 │ Canada │ Class II │ 2025-01-15 │ On Track │ └──────────────────┴──────────┴────────────┴─────────────┴──────────┘ 1 submission at risk: WoundScan Pro EU — NB engagement not confirmed.
| Document | Content | |----------|---------| | fda-submission-guide.md | FDA pathways, requirements, review process | | eu-mdr-submission-guide.md | MDR classification, technical documentation, clinical evidence | | global-regulatory-pathways.md | Canada, Japan, China, Australia, Brazil requirements | | iso-regulatory-requirements.md | ISO 13485, 14971, 10993, IEC 62304, 62366 requirements |
| KPI | Target | Calculation | |-----|--------|-------------| | First-time approval rate | >85% | (Approved without major deficiency / Total submitted) × 100 | | On-time submission | >90% | (Submitted by target date / Total submissions) × 100 | | Review cycle compliance | >95% | (Responses within deadline / Total requests) × 100 | | Regulatory hold time | <20% | (Days on hold / Total review days) × 100 |
| Skill | Integration Point | |-------|-------------------| | mdr-745-specialist | Detailed EU MDR technical requirements | | fda-consultant-specialist | FDA submission deep expertise | | quality-manager-qms-iso13485 | QMS for regulatory compliance | | risk-management-specialist | ISO 14971 risk management |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | fail→pass | 17,028 | 24,310 | +43% | 1 | 1 | 0% | 3,116 | 9,054 | +191% | 0 | 0 | — |
case-09 | fail→pass | 13,935 | 2,675 | -81% | 1 | 1 | 0% | 1,208 | 5,320 | +340% | 0 | 0 | — |
case-20 | fail→pass | 13,532 | 4,958 | -63% | 1 | 1 | 0% | 2,533 | 5,689 | +125% | 0 | 0 | — |
case-01 | fail→fail | 35,671 | 36,031 | +1% | 1 | 1 | 0% | 6,234 | 11,023 | +77% | 0 | 0 | — |
case-07 | pass→pass | 16,920 | 18,337 | +8% | 1 | 1 | 0% | 2,729 | 7,731 | +183% | 0 | 0 | — |
case-08 | fail→pass | 20,298 | 28,939 | +43% | 1 | 1 | 0% | 3,121 | 10,167 | +226% | 0 | 0 | — |
case-02 | fail→fail | 18,906 | 25,264 | +34% | 1 | 1 | 0% | 3,670 | 9,744 | +166% | 0 | 0 | — |
case-03 | fail→fail | 14,899 | 15,663 | +5% | 1 | 1 | 0% | 2,703 | 7,676 | +184% | 0 | 0 | — |
case-04 | pass→pass | 10,117 | 9,852 | -3% | 1 | 1 | 0% | 1,929 | 6,755 | +250% | 0 | 0 | — |
case-05 | pass→pass | 13,368 | 12,399 | -7% | 1 | 1 | 0% | 2,358 | 6,791 | +188% | 0 | 0 | — |
case-06 | pass→pass | 12,184 | 12,402 | +2% | 1 | 1 | 0% | 2,298 | 7,068 | +208% | 0 | 0 | — |
case-10 | pass→pass | 4,398 | 3,838 | -13% | 1 | 1 | 0% | 804 | 5,460 | +579% | 0 | 0 | — |
case-11 | pass→pass | 10,833 | 7,980 | -26% | 1 | 1 | 0% | 1,941 | 6,213 | +220% | 0 | 0 | — |
case-12 | pass→pass | 17,254 | 20,380 | +18% | 1 | 1 | 0% | 3,291 | 8,294 | +152% | 0 | 0 | — |
case-13 | pass→pass | 18,304 | 19,854 | +8% | 1 | 1 | 0% | 3,311 | 8,375 | +153% | 0 | 0 | — |
case-14 | fail→fail | 14,890 | 21,268 | +43% | 1 | 1 | 0% | 2,707 | 8,675 | +220% | 0 | 0 | — |
case-15 | fail→pass | 20,231 | 27,524 | +36% | 1 | 1 | 0% | 3,954 | 9,916 | +151% | 0 | 0 | — |
case-16 | fail→fail | 16,600 | 22,152 | +33% | 1 | 1 | 0% | 2,925 | 8,813 | +201% | 0 | 0 | — |
case-17 | pass→pass | 7,381 | 5,937 | -20% | 1 | 1 | 0% | 1,383 | 5,843 | +322% | 0 | 0 | — |
case-18 | fail→fail | 15,156 | 13,224 | -13% | 1 | 1 | 0% | 2,719 | 7,329 | +170% | 0 | 0 | — |
case-19 | fail→pass | 11,996 | 12,366 | +3% | 1 | 1 | 0% | 2,170 | 7,071 | +226% | 0 | 0 | — |
case-22 | fail→fail | 26,288 | 19,470 | -26% | 1 | 1 | 0% | 5,873 | 9,072 | +54% | 0 | 0 | — |
case-23 | fail→fail | 16,474 | 20,242 | +23% | 1 | 1 | 0% | 3,145 | 8,293 | +164% | 0 | 0 | — |
case-24 | fail→fail | 23,445 | 27,935 | +19% | 1 | 1 | 0% | 5,004 | 10,781 | +115% | 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. 24 cases were attempted. The headline lift of +25 percentage points is the difference between those two pass rates over the 24 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.