mdr-745-specialist

Overview of mdr-745-specialist skill

What mdr-745-specialist is for

mdr-745-specialist is a compliance-focused AI skill for working through EU MDR 2017/745 questions around medical device classification, technical documentation, clinical evidence, PMS, PMCF, PSUR timing, UDI/EUDAMED preparation, and notified body readiness. It is best suited for regulatory affairs, quality, product, clinical, and engineering teams that need a structured MDR review rather than a casual “summarize the regulation” answer.

The real job-to-be-done is practical: turn device facts into an MDR classification rationale, identify technical file gaps against Annex II/III, assess clinical evidence expectations under Annex XIV, and produce action-oriented compliance review outputs.

Best-fit users and review scenarios

Use this skill when you need help with:

• MDR Annex VIII classification, including active devices and software considerations
• Technical documentation structure for Annex II and Annex III
• Clinical evidence planning, equivalence reasoning, PMCF, or CER inputs
• PMS and PSUR cadence planning under MDR requirements
• Pre-audit or pre-notified-body gap checks
• Compliance review of a device description, intended purpose, claims, and available evidence

It is especially useful when you already have device-specific context and want the agent to apply a repeatable MDR workflow.

What makes the skill different from a generic prompt

A generic prompt may answer from broad knowledge. The mdr-745-specialist skill adds a curated MDR workflow plus supporting reference files for classification, clinical evidence, and technical documentation. The repository also includes scripts/mdr_gap_analyzer.py, which can generate a structured gap assessment by device name and class. That combination makes the skill more installation-worthy for teams that want consistent review structure, not just one-off regulatory commentary.

Important adoption boundaries

This skill does not replace a regulatory affairs professional, notified body feedback, legal advice, or a formal quality management system. It is strongest for drafting, triage, gap identification, and review preparation. It is weaker if you provide only a product name without intended purpose, patient population, duration of contact, invasiveness, active/software behavior, claims, or current evidence.

How to Use mdr-745-specialist skill

mdr-745-specialist install and file path

Install from the GitHub skill directory with:

npx skills add alirezarezvani/claude-skills --skill mdr-745-specialist

The upstream skill lives at:

ra-qm-team/skills/mdr-745-specialist

After install, read SKILL.md first, then inspect these support files:

• references/mdr-classification-guide.md
• references/clinical-evidence-requirements.md
• references/technical-documentation-templates.md
• scripts/mdr_gap_analyzer.py

These files matter because the skill’s value is in its MDR-specific decision structure, not just the trigger description.

Inputs that produce better MDR answers

For strong mdr-745-specialist usage, provide device facts in a regulatory format. At minimum, include:

• Device name and intended medical purpose
• User, patient population, and clinical setting
• Whether it is invasive, implantable, active, software-driven, or sterile
• Contact duration and body system interaction
• Claims, indications, contraindications, and risk profile
• Existing classification assumptions
• Available evidence: testing, clinical data, literature, PMS, complaints
• Target output: classification rationale, gap list, technical file outline, CER/PMCF plan, or PSUR review

A weak prompt is: “Classify my device under MDR.”
A stronger prompt is: “Use mdr-745-specialist for Compliance Review. Classify a cloud-based software tool that analyzes ECG data to flag possible arrhythmias for clinician review. It does not provide autonomous treatment decisions. Include Annex VIII reasoning, software classification considerations, assumptions, missing facts, and conformity assessment implications.”

Suggested workflow for compliance review

A practical mdr-745-specialist guide workflow:

1. Start with classification using references/mdr-classification-guide.md.
2. Ask for assumptions and missing facts before accepting the class.
3. Map the likely class to documentation expectations using references/technical-documentation-templates.md.
4. Review clinical evidence expectations using references/clinical-evidence-requirements.md.
5. Run or review the gap analyzer if you want a structured checklist-style output.
6. Convert the result into a remediation plan with owners, evidence needed, and priority.

For the script, inspect usage in scripts/mdr_gap_analyzer.py. The source examples include commands such as:

python mdr_gap_analyzer.py --device "Device Name" --class IIa

and:

python mdr_gap_analyzer.py --interactive

Prompt pattern that invokes the skill well

Use direct role and output instructions:

“Use mdr-745-specialist to perform an EU MDR 2017/745 compliance review for the following device. First identify missing facts. Then provide classification rationale, applicable MDR concepts, technical documentation gaps, clinical evidence expectations, PMS/PMCF/PSUR implications, and a prioritized remediation roadmap. Mark assumptions clearly and avoid final legal conclusions.”

This framing improves output quality because it asks the agent to separate facts, assumptions, MDR reasoning, and next actions.

mdr-745-specialist skill FAQ

Is mdr-745-specialist suitable for beginners?

Yes, if the beginner has device details and wants structured guidance. The references explain MDR classification factors, evidence hierarchy, and technical documentation structure. However, beginners should treat outputs as educational and review-ready drafts, not final regulatory determinations.

When should I not use this skill?

Do not use it as the sole authority for final classification, CE marking strategy, notified body submissions, legal interpretation, or high-risk clinical decisions. Also avoid using it when your device description is too vague. MDR classification depends heavily on intended purpose, invasiveness, duration, body contact, active function, and claims.

How does it help compared with reading MDR directly?

Reading MDR directly is authoritative but time-consuming and difficult to operationalize. The mdr-745-specialist skill translates common MDR work into workflows: classification factors, Annex II/III file structure, clinical evidence planning, PMS expectations, and gap analysis. It helps you ask the right questions before escalating to formal regulatory review.

Does it support software medical devices?

Yes, the repository references software classification, including MDCG 2019-11-style considerations in the classification guide. For software, provide intended purpose, clinical decision impact, data inputs, user role, whether outputs drive diagnosis or treatment, and whether a clinician can independently verify the result.

How to Improve mdr-745-specialist skill

Improve mdr-745-specialist results with stronger facts

The most common failure mode is an overconfident answer from incomplete facts. Improve results by supplying a device fact table before asking for conclusions. Include intended purpose, claims, mechanism of action, patient contact, duration, active/software status, accessories, sterile/measuring function, and markets. Ask the skill to list missing facts before classifying.

Ask for rationale, not just conclusions

For better mdr-745-specialist for Compliance Review outputs, require traceable reasoning:

• “State the classification rule considered.”
• “Explain why alternative classes may or may not apply.”
• “Separate confirmed facts from assumptions.”
• “List evidence needed to defend the conclusion.”
• “Identify notified body implications.”

This makes the output more useful for internal review and easier for a regulatory lead to challenge.

Iterate after the first output

The first response should usually be treated as triage. Follow up with targeted prompts:

• “Revise the classification if the device has long-term body contact.”
• “Convert the gap list into Annex II/III technical file sections.”
• “Turn the clinical evidence gaps into a CER and PMCF evidence plan.”
• “Prioritize gaps for a notified body submission in the next six months.”

Iteration is valuable because MDR conclusions often change when new facts are added.

Extend the skill for your organization

Teams can improve adoption by adding internal templates, device-family examples, risk management links, SOP references, and preferred output formats. Useful additions might include a company technical file index, CER template, PMS plan template, UDI checklist, or a decision log format. Keep local extensions separate from authoritative MDR sources, and label company practice versus regulatory requirement clearly.