Regulatory & Compliance RegulatoryRAMedTech

Regulatory Affairs Head

Senior Regulatory Affairs Manager for HealthTech and MedTech.

$skillfish add ~skills/regulatory-affairs-head

Real Examples Get Started Models & Context

Real Conversations

See how the Regulatory Affairs Head skill responds to actual requests. These are representative examples of what your AI agent will produce.

You

We're a UK-based MedTech startup with a Class IIb AI diagnostic device. We want to sell in the EU, US, and UK. We need a regulatory strategy. What's the most efficient pathway and sequencing?

Agent

Global Regulatory Strategy — Class IIb AI Diagnostic (EU/US/UK)

Recommended sequencing: EU MDR → FDA PMA → UKCA

Rationale: EU MDR is the most stringent and slowest pathway. Completing it first means your clinical data, QMS, and technical documentation are already at the highest standard, making FDA and UKCA submissions faster.

Pathway 1: EU MDR (CE Mark)

Class IIb → Notified Body required (TÜV, BSI, Dekra)
Clinical evidence: Per Art. 61 + MEDDEV 2.7/1 rev. 4. For AI diagnostic: prospective clinical study likely required
AI/ML specific: MDCG 2021-6 guidance on AI classification applies
Timeline: 24–36 months from start
Cost estimate: €300K–600K (Notified Body fees + clinical study + regulatory resource)

Pathway 2: FDA (US)

Class IIb AI diagnostic → likely PMA (Premarket Approval)
Requires: valid scientific evidence (prospective study), manufacturing site inspection (21 CFR Part 820)
AI/ML: FDA's Predetermined Change Control Plan guidance applies if model updates planned
Can leverage EU clinical data if study designed to FDA standards (US site inclusion, FDA-accepted endpoints)
Breakthrough Device Designation: Apply early if your device addresses an unmet need — gets you FDA interaction throughout development
Timeline: 18–24 months post-EU (leveraging EU clinical data)

Pathway 3: UKCA (Great Britain)

Class IIb: UKCA mark required via UK Approved Body
CE marks accepted in GB until June 30, 2028 (extended transition period)
Recommendation: Get CE mark first; CE-to-UKCA conversion is substantially faster
Timeline: 6–12 months post-CE

Parallel workstreams (start now regardless of sequence):

Workstream	Start	Why
ISO 13485 QMS certification	Now	Required for all three pathways
ISO 14971 risk management file	Now	Required for all three pathways
Clinical study protocol design	Now	Must satisfy EU + FDA standards simultaneously
IEC 62304 software lifecycle	Now	Mandatory for AI SaMD everywhere
EUDAMED registration	Month 3	Required for MDR — long lead time

Total investment estimate (3-market strategy): €1M–2M over 36–48 months, dominated by clinical study and Notified Body/FDA fees.

Want me to build the detailed project plan with milestones and a Notified Body selection shortlist?

You

We have a CE-marked Class IIa SaMD (patient monitoring app). We want to release a new software version that adds an AI-based deterioration prediction feature. Does this require a new regulatory submission or can we use our existing CE mark?

Agent

Regulatory Impact Assessment — Software Update with AI Feature

This is one of the most common and consequential questions in SaMD regulation. The answer depends on whether the change is significant under MDR Art. 75.

Step 1: Assess the change against MDR Art. 75 significance criteria

A change is significant if it:

Changes the intended purpose or indication for use
Changes the risk classification
Changes performance specifications in a way that could affect safety
Affects the clinical benefits or introduces new risks not covered in the existing clinical evaluation

Your change: Adding AI-based deterioration prediction

Question	Assessment
Does it change intended purpose?	Likely YES — you're adding a new clinical function (deterioration prediction vs. current monitoring)
Does it change risk classification?	Possibly — prediction of deterioration could elevate from IIa to IIb (Rule 11 analysis required)
Does it affect clinical evaluation?	YES — new AI algorithm requires new clinical validation data
Is it covered by existing CE mark?	No — this is a new intended use not covered in original Technical Documentation

Conclusion: This is almost certainly a significant change requiring Notified Body notification and likely a new Technical Documentation supplement or full re-submission.

Step 2: Notify your Notified Body immediately Under MDR Art. 75, you must notify your NB of planned significant changes before implementation. Do not release the update without NB review.

Submit a Change Notification including:

Description of the change
Rationale for significance assessment
Impact on risk management file
Clinical data supporting the new AI feature
Updated SoA (Statement of Applicability) if design controls changed

Step 3: Clinical evidence for the AI feature The new deterioration prediction algorithm requires:

Algorithm validation data: sensitivity, specificity, positive/negative predictive value vs. clinician assessment
Performance data in the intended patient population
Evidence that the AI output is appropriately qualified (e.g., decision support only, not autonomous action)

Step 4: Consider a Predetermined Change Control Plan (PCCP) If you anticipate regular AI model updates, work with your NB to establish a PCCP that pre-approves the methodology for future updates, reducing submission burden for iterative improvements.

Timeline for NB review of significant change: Typically 3–6 months.

Risk of releasing without NB approval: Non-conformant CE mark, potential MDR Art. 87 vigilance obligation if issues occur, liability exposure.

Want me to draft the Change Notification document for your Notified Body?

Get Started

How to install and use this skill in your preferred environment.

Skills are designed for AI coding agents (Claude Code, Cursor, Windsurf) and IDE-based workflows where the agent can read files, run scripts, and act on your codebase. Web-based AI can use the knowledge and frameworks, but won't have tool access.

Models & Context

Which AI models and context windows work best with this skill.

Recommended Models

Best

Claude Opus 4 Claude Sonnet 4 GPT-4.1 Gemini 2.5 Pro Grok 3 Kimi K2

Good

Claude Haiku 4.5 GPT-4.1 mini Gemini 2.5 Flash Grok 3 mini

Larger models produce more detailed, production-ready outputs.

Context Window

This skill's SKILL.md is typically 3–10 KB — fits in any modern context window.

8K Skill only

32K+ Skill + conversation

100K+ Skill + references + codebase

All current frontier models (Claude, GPT, Gemini) support 100K+ context. Use the full window for complex multi-service work.

Pro tips for best results

1

Be specific

Include numbers — users, budget, RPS — so the skill can size the architecture.

2

Share constraints

Compliance needs, team size, and existing stack all improve the output.

3

Iterate

Start with a high-level design, then ask follow-ups for IaC, cost analysis, or security review.

4

Combine skills

Pair with companion skills below for end-to-end coverage.

Ready to try Regulatory Affairs Head?

Install the skill and start getting expert-level guidance in your workflow — any agent, any IDE.

$skillfish add ~skills/regulatory-affairs-head

← Browse all 169 skills