fda-qsr-audit-prep

$npx mdskill add alirezarezvani/claude-skills/fda-qsr-audit-prep

**Command:** `/compliance-os:fda-qsr-audit-prep <scope>`

SKILL.md

.github/skills/fda-qsr-audit-prepView on GitHub ↗
---
name: "fda-qsr-audit-prep"
description: "/compliance-os:fda-qsr-audit-prep [scope] — FDA 21 CFR 820 (QSR / QMSR) audit 6-question forcing interrogation. Post-Feb 2026 substantially harmonized with ISO 13485. Use before annual internal QSR audit, pre-FDA-inspection readiness, or Form 483 response."
---

# /compliance-os:fda-qsr-audit-prep — FDA QSR Forcing Questions

**Command:** `/compliance-os:fda-qsr-audit-prep <scope>`

The FDA QSR auditor pressure-tests any US medical-device QSR work. Six questions before any internal audit, FDA inspection, Form 483 response, or recall decision.

## When to Run

- Before annual internal QSR audit
- Before pre-FDA-inspection readiness review (any device commercially distributed in US)
- After receiving Form 483 observations
- After Warning Letter receipt
- After MDR-reportable event
- Before recall decision (voluntary vs FDA-initiated)
- Before submitting 510(k) / PMA (where QSR posture affects approval timeline)

## The Six QSR Questions

### 1. Show me the complaint files from the last quarter — and the corresponding MDR reports.
**21 CFR 820.198 + 21 CFR 803 — most-cited FDA inspection area.**
- Complaint log complete: who / what / when / device / batch
- Investigation closure within reasonable timeline
- MDR-reporting decision tree applied: death OR serious injury OR malfunction-that-could-cause = MDR
- 30-day timeline for most MDR reports; 5 days for certain serious events
- Complaint trending input to management review

### 2. When was process validation (IQ/OQ/PQ) last revalidated per 21 CFR 820.75?
**Cross-walks ISO 13485 Clause 7.5.6 (substantially harmonized post-Feb 2026).**
- Initial validation at process introduction
- Revalidation triggers: process / equipment / material change OR periodic schedule
- Statistical techniques per 21 CFR 820.250 where applicable
- Cross-check with cs-cqm-iso13485 for ISO 13485 alignment

### 3. Show me the DHRs for products commercially distributed in last 2 years.
**21 CFR 820.180 — 2-year retention from commercial distribution; check sampling for completeness.**
- Device History Record (DHR) for each unit/lot/batch
- Must include: dates of manufacture, quantity manufactured, quantity released, acceptance records, primary identification label, device identification, control number
- Sample stratified by product class
- Verify DHR closeness to DHF (design history file)

### 4. Show me CAPAs from the last 6 months with effectiveness verification.
**21 CFR 820.100 = ISO 13485 8.5.2 substantially harmonized.**
- Root cause analysis depth (5 Why minimum)
- Effectiveness verification = measurable evidence, not "we updated the procedure"
- Containment / correction / corrective action distinction documented
- Closure approval by appropriate authority
- Aging CAPAs > 90 days flagged

### 5. Show me labeling (21 CFR 801) review for the most recent product launch.
**FDA-specific overlay not in ISO 13485.**
- Labeling per 21 CFR 801 requirements
- For specific device types: also 21 CFR 800 series sectoral overlays
- UDI (Unique Device Identification) per 21 CFR 830
- Promotional materials reviewed for accuracy + non-misleading

### 6. If a Form 483 was issued in the last 3 years, show me the closure status.
**Form 483 = FDA observation; not equivalent to ISO nonconformity.**
- Response within 15 working days
- Each observation has documented corrective + preventive action with timeline
- Effectiveness verification evidence
- For Warning Letters: separate response track + potentially FDA meeting

## Workflow

```bash
# 1. QSR compliance posture
python ${CLAUDE_PLUGIN_ROOT}/vendor/ra-qm-team/fda-consultant-specialist/scripts/qsr_compliance_checker.py compliance_state.json

# 2. FDA submission tracking (510(k) / PMA / IDE)
python ${CLAUDE_PLUGIN_ROOT}/vendor/ra-qm-team/fda-consultant-specialist/scripts/fda_submission_tracker.py submissions.json

# 3. HIPAA overlap (if connected device handles PHI)
python ${CLAUDE_PLUGIN_ROOT}/vendor/ra-qm-team/fda-consultant-specialist/scripts/hipaa_risk_assessment.py phi_inventory.json

# 4. Mock FDA inspection
python ../../skills/compliance-os/scripts/audit_simulator.py fda_qsr_scope.json
```

## Output Format

```markdown
# FDA QSR Audit Prep: <scope>
**Date:** YYYY-MM-DD

## The Decision Being Made
[programme-plan | inspection-readiness | 483-response | MDR-decision | recall]

## Complaint + MDR Posture
- Complaints last quarter: N
- MDR-reportable events: M
- MDR reports filed within timeline: % (target 100%)
- Complaint trending review at management level: yes/no

## Process Validation Status (21 CFR 820.75)
- Validations on schedule: %
- Stale validations: <list>
- Statistical techniques applied: yes/no per process

## DHR Completeness (21 CFR 820.180)
- DHRs sampled: N
- Completeness rate: %
- 2-year retention compliant: yes/no
- Stratified by product class: yes/no

## CAPA Health (21 CFR 820.100)
- CAPAs sampled: N
- Root cause analysis depth: adequate/inadequate
- Effectiveness verification: complete/incomplete
- Aging CAPAs > 90 days: N

## Labeling (21 CFR 801)
- Recent products reviewed: <list>
- Labeling accurate + non-misleading: yes/no
- UDI compliance per 21 CFR 830: yes/no

## Form 483 / Warning Letter History
- Form 483s last 3 years: N (each: closed/in-progress)
- Warning Letters last 5 years: N (each: closed/in-progress)
- Pattern across observations: <thematic>

## ISO 13485 Cross-Walk (post-Feb 2026 harmonization)
- ISO 13485 audit findings: <link to cs-cqm-iso13485 output>
- FDA-specific overlays remaining: labeling + complaint handling + MDR reporting + recall procedures
- Cross-framework reuse: % of evidence shared

## Verdict
🟢 INSPECTION-READY | 🟡 GAPS-IDENTIFIED | 🔴 NOT-READY

## Top 3 Actions
[3 concrete next steps with owner + FDA-cited timeline (15 days / 30 days / etc.)]

## Outside Counsel Required
[For Warning Letter response, recall decisions, or 510(k) / PMA strategy disputes]
```

## Routing

- `/compliance-os:compliance-readiness` — for multi-framework view
- `/compliance-os:iso13485-audit-prep` — for ISO 13485 cross-walk pair (substantially harmonized)
- `/compliance-os:gdpr-audit-prep` — if connected device handles personal data
- `/c-level-agents:gc-review` — for Warning Letter response coordination

## Related

- Agent: [`cs-fda-qsr-auditor`](../../agents/cs-fda-qsr-auditor.md)
- Skill: [`fda-consultant-specialist`](${CLAUDE_PLUGIN_ROOT}/vendor/ra-qm-team/fda-consultant-specialist/SKILL.md)
- Adjacent: `../iso13485-audit-prep/`, `../compliance-readiness/`

---

**Version:** 1.0.0

More from alirezarezvani/claude-skills

SkillDescription
a11y-auditAccessibility audit skill for scanning, fixing, and verifying WCAG 2.2 Level A and AA compliance across React, Next.js, Vue, Angular, Svelte, and plain HTML codebases. Use when auditing accessibility, fixing a11y violations, checking color contrast, generating compliance reports, or integrating accessibility checks into CI/CD pipelines.
ab-test-setupWhen the user wants to plan, design, or implement an A/B test or experiment. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "conversion experiment," "statistical significance," or "test this." For tracking implementation, see analytics-tracking.
ad-creativeWhen the user needs to generate, iterate, or scale ad creative for paid advertising. Use when they say 'write ad copy,' 'generate headlines,' 'create ad variations,' 'bulk creative,' 'iterate on ads,' 'ad copy validation,' 'RSA headlines,' 'Meta ad copy,' 'LinkedIn ad,' or 'creative testing.' This is pure creative production — distinct from paid-ads (campaign strategy). Use ad-creative when you need the copy, not the campaign plan.
adversarial-reviewerAdversarial code review that breaks the self-review monoculture. Use when you want a genuinely critical review of recent changes, before merging a PR, or when you suspect Claude is being too agreeable about code quality. Forces perspective shifts through hostile reviewer personas that catch blind spots the author's mental model shares with the reviewer.
aeoAnswer Engine Optimization (AEO) skill — optimize content to be cited by AI language models (ChatGPT, Perplexity, Claude, Gemini, Mistral) as authoritative sources. Distinct from SEO — AEO optimizes for citation in LLM-generated responses, not search rankings. Use when planning content for AI-first search audiences, auditing existing content for E-E-A-T signals, tracking which pages get cited by which LLMs, or building a citation-friendly content strategy. Triggers — 'AEO audit', 'optimize for ChatGPT', 'get cited by Perplexity', 'LLM citation strategy', 'answer engine optimization', 'content for AI search', 'E-E-A-T audit'. Output is a markdown audit report (default) or JSON for pipeline integration. Stdlib-only Python tools.
agent-decision-receiptsMint a tamper-evident, post-quantum-signed receipt for a consequential agent action (deploy, delete, pay, grant-access, model decision) so it can be verified later from the certificate alone. Use when an autonomous agent takes a side-effecting action that may need to be proven later, or when satisfying EU AI Act Article 12 record-keeping. Three decisions: whether an action needs a receipt, minting it, verifying it. Signing is delegated to the open-source OpenAgentOntology package. Not after-the-fact log analysis; not a hosted notary; not a legal opinion.
agent-designerUse when the user asks to design a multi-agent system, pick an orchestration pattern (supervisor/swarm/pipeline), generate tool schemas for agents, or evaluate agent execution logs for cost, latency, and failure bottlenecks. Examples: 'design an agent architecture for research automation', 'generate Anthropic tool schemas from these tool descriptions', 'analyze these agent run logs for bottlenecks'. NOT for Claude Code workflow files (use workflow-builder) or single-agent prompt design (use agent-workflow-designer).
agent-harnessTurn any domain folder of skills into a bounded agentic loop: compile a goal into a verifiable task plan, execute tasks with the domain's own tools, verify every task with machine-run checks, retry with caps, escalate to a human when budgets exhaust, and refuse to close until everything is verified or explicitly waived. Use when you want an agent or subagent to pick up a goal and drive it to a verified close across one of this repo's 18 domains ('run this goal through the engineering harness', 'set up an agentic loop for marketing work', 'make the finance domain self-verifying'). NOT for authoring Claude Code Workflow-tool .js scripts (workflow-builder), N-agent tournaments on one task (agenthub), single-file metric optimization (autoresearch-agent), or discovering published loop recipes (loop-library).
agent-launcher-orchestratorUse when a user wants to build, launch, grade, or schedule a Claude Managed Agent (CMA) in their own Anthropic account — "build me an agent", "launch this as a managed agent", "run this on a schedule", "grade my agent against a rubric", "set up a nightly worker". Reads the per-session goal (./my-agent/goal.json), routes deterministically to one of five phase sub-skills (interview → stage-launch → grade-iterate → run-without-you → wrap-up) via goal_router.py, and compiles the goal+phase into an execution shape (single-pass workflow / bounded grade→iterate loop / recurring cron deployment loop) via loop_compiler.py. Forks context so heavy intake (build sheets, payloads, eval cases) stays out of the parent thread. All launches are emitted as BYOK curl the user runs with their own key; no tool makes API calls. Inspired by anthropics/launch-your-agent (Apache-2.0). Distinct from engineering/agent-harness (generic domain loop) and engineering/write-a-skill (authors Claude Code skills, not CMAs).
agent-memoryUse when a project's CLAUDE.md has grown past what anyone reads and you want the agent to learn durable facts from its own sessions instead — or when asking why the agent keeps re-learning the same correction, why a remembered rule is wrong, or where a memory line came from. Implements a four-tier store (L0 transcripts / L1 candidates / L2 project context / L3 stable persona) where promotion is earned by recurrence across sessions and days, never by one confident statement, and nothing reaches a committed file without a human adopting it.