bundled-skills/fda-medtech-compliance-auditor/SKILL.md
Expert AI auditor for Medical Device (SaMD) compliance, IEC 62304, and 21 CFR Part 820. Reviews DHFs, technical files, and software validation.
npx skillsauth add FrancoStino/opencode-skills-antigravity fda-medtech-compliance-auditorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill transforms your AI assistant into a specialized MedTech Compliance Auditor. It focuses on Software as a Medical Device (SaMD) and traditional medical equipment regulations, including 21 CFR Part 820 (Quality System Regulation), IEC 62304 (Software Lifecycle), ISO 13485, and ISO 14971 (Risk Management).
@fda-medtech-compliance-auditor and provide the document you wish to review.Scenario: A CAPA was opened for a software defect in a Class II device. The documented root cause is “developer error — unclear requirements.” The corrective action is developer retraining.
Finding:
FDA AUDIT FINDING
Severity: Major
Citation: 21 CFR 820.100(a)(2) / IEC 62304 Section 5.1
Analysis:
"Developer error" is a symptom, not a root cause. Retraining alone is
a known red flag for FDA inspectors and will not withstand scrutiny.
The true root cause lies in the software requirements engineering
process itself — not an individual.
Required Actions:
1. Perform a 5-Whys or Fishbone analysis targeting the requirements
gathering and review process.
2. Update the SRS (Software Requirements Specification) and the
corresponding process SOP.
3. Document an effectiveness check with a measurable criterion
(e.g., zero requirements-related defects in next 3 releases).
4. Do not close the CAPA on retraining alone.
tools
Authorized security assessment of LLM applications and AI agents: prompt injection, tool abuse, RAG exposure, memory poisoning, system-prompt extraction, and agent-compliance engineering per OWASP LLM/ASI Top 10.
development
Builds two parameterized UI modes—流光溢彩白 (iridescent white) and 五彩斑斓黑 (colorful black)—with OKLCH, WebGL/CSS fallback, vision gating, screenshot QA, and total/per-color intensity reports. Use when a UI request names either mode or needs measured color parameters.
tools
Delegate coding tasks to the Kimi Code CLI (`kimi`) only when the user explicitly requests it, while the orchestrator retains review and landing responsibility.
development
Front-end JavaScript reverse engineering: locate signature chains, analyze encrypted request parameters, sample runtime behavior, and reproduce logic locally in Node for evidence-based output.