regulatory-affairs-head

I checked out your regulatory-affairs-head skill and noticed you're building something that bridges the gap between technical implementation and compliance requirements—that's a tough problem space that most skills don't even attempt. The 86 score feels right; there's solid foundational work here, but I'm curious how you're handling the inherent tension between keeping compliance guidance accessible and technically precise.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 86/100, solid B territory. This is based on Anthropic's skill best practices rubric. Your Writing Style is your strongest area (9/10)—the voice is consistent and objective throughout. The weakest pull is Spec Compliance (12/15), mainly because your description could use more trigger phrases to help discovery.

What's Working Well

• Progressive disclosure structure is clean — You've properly delegated detailed guidance to reference files (eu-mdr-submission-guide.md, fda-submission-guide.md) while keeping SKILL.md as an overview. That's the right move for a complex domain.

• Strong trigger terms for discoverability — Your description nails specific regulatory workflows: "regulatory strategy," "submission planning," "risk assessment." These will actually help the right people find this skill.

• Practical decision framework — The "Regulatory Decision Framework" with numbered steps (1. Impact Assessment → 2. Pathway Selection → 3. Timeline Planning) gives users a clear entry point. That structure works.

• Comprehensive pathway coverage — Addressing both FDA and EU MDR submissions shows you understand the scope. That's not trivial.

The Big One

You're referencing files that don't exist — This is the heavyweight issue. Your SKILL.md references iso-regulatory-requirements.md, global-regulatory-pathways.md, and escalation-procedures.md, plus asset directories (assets/communication-templates/, assets/submission-checklists/, etc.), but they're not in the repo.

This isn't just a documentation gap—it breaks functionality. When someone activates this skill and Claude tries to follow your reference links, they hit dead ends. That tanks the utility score hard.

Fix: Either create the missing reference files or be honest about what exists. If you want to keep references minimal for now, swap the specific file names for something like "See the references/ directory for regulatory pathway guides and templates." You can expand the reference library later without breaking the skill.

Other Things Worth Fixing

1. Add a table of contents — Your SKILL.md is 128 lines and the FDA/EU reference files are around 128 lines each. Files that long need a TOC at the top. Just add ## Contents after the frontmatter with links to major sections. That's +2-3 points for navigation.

2. Concrete examples would help — You mention templates and checklists but don't show samples. Add something like: "Example Regulatory Update: Subject: [Product]-[Submission Type] Status Update, Status: Under Review..." This gives people a mental model of what output actually looks like.

3. Trim the header labels — "Key Activities" and "Team Coordination Protocol" are a bit redundant. You could just use the list format directly: - Regulatory pathway analysis instead of wrapping it. Saves tokens and reads cleaner.

Quick Wins

• Create or remove missing references (biggest impact—fixes the high-severity issue)
• Add TOC to long files (SKILL.md, FDA/EU guides over 100 lines)
• Swap generic headers for direct lists (cleaner, fewer tokens)
• Add 1-2 concrete input/output examples (makes utility real)

Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.