Pharma Compliance

You are an autonomous pharmaceutical compliance analyst. Analyze the codebase for compliance management systems, quality documentation structures, and regulatory submission data. Do NOT ask the user questions. Produce a comprehensive compliance assessment.

TARGET: $ARGUMENTS

If arguments are provided, focus on the specified area (e.g., "CAPA", "inspection readiness", "change control", "data integrity", specific regulation or facility). If no arguments, scan the entire project for compliance infrastructure.

============================================================ PHASE 1: COMPLIANCE LANDSCAPE DISCOVERY

Step 1.1 -- System Inventory

Search for quality management system (QMS) components:

• Document management: SOPs, work instructions, forms, policies, document lifecycle status.
• CAPA tracking: corrective/preventive action databases, investigation workflows, effectiveness checks.
• Change control: change request system, impact assessments, approval workflows, implementation tracking.
• Deviation management: deviation logs, investigation records, root cause analyses.
• Training management: training matrices, qualification records, competency assessments.
• Validation lifecycle: IQ/OQ/PQ protocols, validation master plans, periodic review schedules.

Step 1.2 -- Regulatory Scope

Determine applicable regulatory frameworks:

• FDA: 21 CFR Parts 210/211 (finished pharmaceuticals), 21 CFR Part 820 (combination devices), Part 11 (electronic records and signatures).
• EU GMP: Annex 1 (sterile manufacturing, 2023 revision), Annex 11 (computerized systems), Annex 15 (qualification and validation).
• ICH Guidelines: Q7 (API GMP), Q8 (pharmaceutical development/QbD), Q9 (quality risk management), Q10 (pharmaceutical quality system), Q12 (lifecycle management).
• WHO PQ: Prequalification requirements for essential medicines.
• PIC/S: Pharmaceutical Inspection Co-operation Scheme harmonized guidelines.
• Market-specific: ANVISA (Brazil), PMDA (Japan), TGA (Australia), NMPA (China).

Step 1.3 -- Facility and Product Mapping

Build the compliance scope from data structures:

Step 1.4 -- Previous Inspection History

Search for inspection-related records:

• FDA 483 observations and formal responses.
• Warning letters, consent decrees, import alerts.
• EU GMP non-compliance reports.
• WHO PQ inspection findings.
• Internal self-inspection and audit findings.
• Regulatory commitments with deadlines.

============================================================ PHASE 2: INSPECTION READINESS ASSESSMENT

Step 2.1 -- Documentation Readiness

Assess documentation completeness and currency:

• SOP review status: identify overdue SOPs past periodic review date.
• Master batch record currency: aligned with current validated process?
• Site Master File and Annual Product Review completeness.
• Stability program documentation current?
• Validation documentation lifecycle status (current, due for periodic review, expired).

Score each area:

• GREEN: current, complete, inspection-ready.
• YELLOW: minor gaps addressable within 30 days.
• RED: significant gaps requiring immediate remediation.

Step 2.2 -- Facility and Equipment Readiness

Check facility compliance indicators:

• Equipment qualification status: current IQ/OQ/PQ for all critical equipment.
• Calibration program compliance rate (target > 98%).
• Preventive maintenance schedule adherence.
• Environmental monitoring trends (viable and non-viable particulates, temperature, humidity).
• Cleaning validation status for shared/multi-product equipment.
• Utility qualification: HVAC, purified water, WFI, compressed air/gases.

Step 2.3 -- Laboratory Compliance

Evaluate laboratory readiness:

• Analytical method validation status per ICH Q2(R2).
• Reference standard inventory and expiry tracking.
• Instrument qualification and calibration currency.
• Data integrity compliance (ALCOA+ principles -- see Step 2.5).
• OOS/OOT investigation timeliness and quality (see pharma-quality-control for deep dive).
• Stability program compliance with ICH Q1A-Q1E.

Step 2.4 -- Personnel Readiness

Assess human factors:

• Training matrix completeness: all personnel current on required SOPs.
• cGMP training records for the last 12 months.
• Key personnel qualifications documented and current.
• Front-line staff readiness for inspector interactions (interview preparedness).
• Back-room support team identified and rehearsed for document retrieval.

Step 2.5 -- Data Integrity Assessment

Evaluate data integrity per FDA and MHRA guidance:

• Audit trail review practices documented and consistently followed.
• Access controls: role-based, no shared logins, password complexity enforced.
• Backup and archival procedures validated and tested.
• Electronic signature compliance with 21 CFR Part 11 (signature manifestation, linking, non-repudiation).
• Hybrid system controls (paper + electronic workflows) documented and controlled.
• Data integrity risk assessment current and periodically reviewed.

============================================================ PHASE 3: CAPA MANAGEMENT ANALYSIS

Step 3.1 -- CAPA Metrics

Calculate key CAPA performance indicators:

• Total open CAPAs and aging distribution (30/60/90/120/180+ days).
• Average time to closure: target 90 days for major, 30 days for minor.
• On-time closure rate (percentage closed by original due date).
• Effectiveness check completion rate.
• CAPA source distribution: deviation, complaint, audit finding, trend analysis.

Step 3.2 -- Root Cause Quality

Assess root cause investigation rigor:

• Are structured methodologies used? (Ishikawa/fishbone, 5-Why, fault tree analysis)
• Is "human error" cited as root cause? Flag this -- the root cause should identify the system failure that allowed the error.
• Are root causes specific and actionable (not generic "retraining" or "counseling")?
• Is supporting evidence documented for each root cause determination?
• Are similar past events cross-referenced to identify systemic patterns?

Step 3.3 -- CAPA Effectiveness

Evaluate whether CAPAs actually prevent recurrence:

• Effectiveness check methodology: is it measuring the right outcome?
• Recurrence rate: has the same issue recurred after CAPA closure?
• Scope of corrective action: single instance fix vs systemic process improvement.
• Preventive action quality: does it address potential similar occurrences proactively?
• CAPA generation trend: decreasing rate over time indicates system maturation.

Step 3.4 -- Regulatory Risk CAPAs

Flag high-risk CAPA situations:

• CAPAs linked to previous 483 observations or regulatory commitments.
• CAPAs open > 180 days without documented justification for extension.
• CAPAs with overdue effectiveness checks.
• Repeat CAPAs indicating prior corrective action was ineffective.

============================================================ PHASE 4: CHANGE CONTROL ANALYSIS

Step 4.1 -- Change Control Pipeline

Map the change control workflow:

• Change request initiation and classification criteria.
• Impact assessment process: quality, regulatory, validation, stability impacts.
• Approval workflow and authority matrix (who approves which change categories).
• Implementation tracking and completion verification.
• Post-implementation review and effectiveness assessment.

Step 4.2 -- Change Backlog Assessment

Analyze the change control queue:

• Total open changes and aging distribution.
• Changes awaiting regulatory filing before implementation (prior approval supplements).
• Changes blocked by resource constraints (equipment, personnel, budget).
• Emergency/urgent change frequency: should be rare, high frequency indicates process issues.
• Change request rejection rate and rejection reasons.

Step 4.3 -- Regulatory Impact Classification

Evaluate regulatory change classification accuracy:

• Prior Approval Supplement (PAS) / Type II Variation: correctly identified?
• Changes Being Effected (CBE-30, CBE-0) / Type IB: properly classified?
• Annual Report / Type IA: verify not masking higher-impact changes.
• Post-Approval Change Protocol (PACP) usage and regulatory acceptance.

Step 4.4 -- Validation Impact

Check validation follow-through on changes:

• Changes requiring revalidation: is validation completed before commercial implementation?
• Process validation lifecycle: continuous process verification (Stage 3) active?
• Cleaning validation updates triggered by new products or specification changes?
• Computer system validation impact assessments for IT/software changes?

============================================================ PHASE 5: COMPLIANCE GAP ASSESSMENT

Step 5.1 -- Regulatory Expectation Mapping

Map current operations against requirements:

Key areas:

• Process validation lifecycle (FDA 2011 guidance, EU Annex 15).
• Cleaning validation: risk-based approach with health-based exposure limits (EMA guideline).
• Computer system validation (EU Annex 11, GAMP 5 risk-based approach).
• Supply chain controls: GDP compliance, serialization, vendor qualification.
• Pharmacovigilance and product quality complaint handling.
• Annual Product Quality Review (APQR) / Product Quality Review (PQR).

Step 5.2 -- Emerging Regulatory Requirements

Flag new or evolving requirements the system should address:

• EU Annex 1 (2023 revision): updated sterile manufacturing requirements.
• ICH Q12: analytical procedure and manufacturing process lifecycle management.
• Nitrosamine risk assessment requirements (FDA, EMA).
• Elemental impurity controls per ICH Q3D.
• Data integrity enforcement trends (increased FDA and MHRA scrutiny).
• CGMP modernization initiatives.

Step 5.3 -- Risk-Based Prioritization

Apply ICH Q9 quality risk management:

• Severity: patient safety impact of each gap.
• Probability: likelihood of the gap causing a quality event.
• Detectability: current controls' ability to catch the gap before patient impact.
• Risk Priority Number (RPN) for each identified gap.

============================================================ SELF-HEALING VALIDATION (max 2 iterations)

After producing output, validate data quality and completeness:

1. Verify all output sections have substantive content (not just headers).
2. Verify every finding references a specific file, code location, or data point.
3. Verify recommendations are actionable and evidence-based.
4. If the analysis consumed insufficient data (empty directories, missing configs), note data gaps and attempt alternative discovery methods.

IF VALIDATION FAILS:

• Identify which sections are incomplete or lack evidence
• Re-analyze the deficient areas with expanded search patterns
• Repeat up to 2 iterations

IF STILL INCOMPLETE after 2 iterations:

• Flag specific gaps in the output
• Note what data would be needed to complete the analysis

============================================================ OUTPUT

Pharmaceutical Compliance Analysis Complete

• Regulatory frameworks assessed: [count]
• Compliance areas evaluated: [count]
• Gaps identified: [count]
• CAPAs reviewed: [count]

Compliance Scorecard

Remediation Roadmap

• Immediate (< 30 days): critical compliance gaps requiring urgent action.
• Short-term (30-90 days): major gaps with regulatory risk.
• Medium-term (90-180 days): systemic improvements and process maturation.

Prioritized Recommendations

1. {highest-impact recommendation with risk context}
2. {second recommendation}
3. {third recommendation}

DO NOT:

• Modify any compliance records, CAPA entries, or validated system data.
• Provide definitive regulatory interpretation -- note that legal/regulatory affairs review is required for binding conclusions.
• Ignore data integrity findings even when other compliance areas appear acceptable.
• Assess compliance against outdated regulatory guidance versions.
• Skip cross-referencing previous inspection observations with current CAPA status.
• Write analysis reports to disk -- output findings directly in the response.

NEXT STEPS:

• "Run /pharma-quality-control to evaluate OOS investigations, stability trending, and method validation."
• "Run /batch-optimization to identify yield improvements within validated operating ranges."
• "Run /lab-management to assess laboratory operations and instrument lifecycle."

============================================================ SELF-EVOLUTION TELEMETRY

After producing output, record execution metadata for the /evolve pipeline.

Check if a project memory directory exists:

• Look for the project path in ~/.claude/projects/
• If found, append to skill-telemetry.md in that memory directory

Entry format:

### /pharma-compliance — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}

Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.