Install any skill in seconds. Free to start, no credit card required.
Get Started Free →GDPR audit-prep playbook: sprint to prepare for a supervisory authority inquiry, DPA audit, or internal review. Use when an audit is scheduled, when readiness gaps surface, or when ROPA (Records of Processing Activities) needs completion.
.claude/skills/borghei-gdpr-audit-prep/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 66% | 0% |
Operational playbook for GDPR audit preparation — whether triggered by a Data Protection Authority (DPA) inquiry, customer-side DPA review, internal compliance audit, or annual self-assessment.
When to use this skill vs. gdpr-dsgvo-expert:
| Situation | Skill applies | |-----------|---------------| | Supervisory authority inquiry received | Yes — start immediately; engage DPO | | Customer DPA audit / questionnaire | Yes — scripts/gdpr_readiness_score.py first | | Annual internal GDPR audit | Yes — standard sprint | | ROPA needs rapid update | Yes — scripts/ropa_completeness_checker.py | | New high-risk processing → DPIA needed | Use ra-qm-team/gdpr-dsgvo-expert for DPIA design |
Week 1: ROPA review + DPO engagement plan
Week 2: Gap remediation (policies, notices, technical)
Week 3: Evidence compilation + walkthroughs
Week 4: Audit week / submissionWeeks 1-2: ROPA update + gap identification
Weeks 3-5: Gap closure (DPAs, notices, security, retention)
Weeks 6-7: Evidence + walkthroughs
Week 8: AuditWeeks 1-2: Inquiry analysis + response strategy + DPO engagement
Weeks 3-8: Targeted evidence collection + remediation
Weeks 9-10: Formal response drafting + legal review
Weeks 11-12: Submission + ongoing dialogueSee references/gdpr-pre-audit-checklist.md for the full pre-audit punch list and references/dpo-engagement-playbook.md for DPO-coordinated audit response.
Every processing activity documented:
Audit gap: ROPA incomplete, stale, or missing for processing activities surfaced during audit.
Before running the audit-prep, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the readiness assessment.
python3 scripts/gdpr_readiness_score.py --config gdpr-controls.yamlpython3 scripts/ropa_completeness_checker.py --ropa ropa.yaml| Script | Purpose | |--------|---------| | scripts/gdpr_readiness_score.py | Score current state per GDPR area; identify gaps | | scripts/ropa_completeness_checker.py | Validate ROPA structure and completeness per Article 30 |
ra-qm-team/gdpr-dsgvo-expert — deep GDPR program managementra-qm-team/audit-prep/compliance-readiness — multi-framework readiness (GDPR + ISO 27001 + SOC 2)ra-qm-team/ccpa-cpra-privacy-expert — US privacy counterpartra-qm-team/audit-prep/ai-act-readiness — EU AI Act overlay for AI processing| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→pass | 15,295 | 13,672 | -11% | 1 | 1 | 0% | 2,317 | 3,907 | +69% | 0 | 0 | — |
case-01 | fail→fail | 19,454 | 15,780 | -19% | 1 | 1 | 0% | 3,029 | 4,260 | +41% | 0 | 0 | — |
case-02 | pass→pass | 23,441 | 18,304 | -22% | 1 | 1 | 0% | 3,727 | 4,628 | +24% | 0 | 0 | — |
case-03 | fail→pass | 17,576 | 16,026 | -9% | 1 | 1 | 0% | 2,898 | 4,306 | +49% | 0 | 0 | — |
case-05 | fail→pass | 16,164 | 14,431 | -11% | 1 | 1 | 0% | 2,461 | 4,065 | +65% | 0 | 0 | — |
case-06 | pass→pass | 14,749 | 12,805 | -13% | 1 | 1 | 0% | 2,512 | 3,712 | +48% | 0 | 0 | — |
case-07 | fail→fail | 11,395 | 7,676 | -33% | 1 | 1 | 0% | 1,657 | 2,890 | +74% | 0 | 0 | — |
case-08 | pass→pass | 10,437 | 7,916 | -24% | 1 | 1 | 0% | 1,628 | 3,089 | +90% | 0 | 0 | — |
case-09 | pass→pass | 7,031 | 7,796 | +11% | 1 | 1 | 0% | 1,095 | 2,957 | +170% | 0 | 0 | — |
case-10 | pass→pass | 15,713 | 15,363 | -2% | 1 | 1 | 0% | 2,547 | 4,136 | +62% | 0 | 0 | — |
case-11 | pass→pass | 17,977 | 16,021 | -11% | 1 | 1 | 0% | 2,717 | 4,180 | +54% | 0 | 0 | — |
case-12 | fail→pass | 15,199 | 13,058 | -14% | 1 | 1 | 0% | 2,610 | 3,752 | +44% | 0 | 0 | — |
case-13 | pass→pass | 15,410 | 14,458 | -6% | 1 | 1 | 0% | 2,312 | 4,067 | +76% | 0 | 0 | — |
case-14 | pass→pass | 14,429 | 12,090 | -16% | 1 | 1 | 0% | 2,382 | 3,806 | +60% | 0 | 0 | — |
case-15 | fail→pass | 13,153 | 9,926 | -25% | 1 | 1 | 0% | 2,061 | 3,411 | +66% | 0 | 0 | — |
case-16 | pass→pass | 27,739 | 30,777 | +11% | 1 | 1 | 0% | 4,513 | 6,734 | +49% | 0 | 0 | — |
case-17 | pass→pass | 8,924 | 6,268 | -30% | 1 | 1 | 0% | 1,332 | 2,625 | +97% | 0 | 0 | — |
case-18 | pass→pass | 19,511 | 16,256 | -17% | 1 | 1 | 0% | 3,068 | 4,349 | +42% | 0 | 0 | — |
case-19 | fail→fail | 16,872 | 19,141 | +13% | 1 | 1 | 0% | 2,661 | 4,639 | +74% | 0 | 0 | — |
case-20 | fail→fail | 25,987 | 28,199 | +9% | 1 | 1 | 0% | 4,147 | 6,262 | +51% | 0 | 0 | — |
case-21 | fail→fail | 20,320 | 22,514 | +11% | 1 | 1 | 0% | 3,182 | 5,035 | +58% | 0 | 0 | — |
case-22 | fail→fail | 25,172 | 29,841 | +19% | 1 | 1 | 0% | 4,583 | 6,866 | +50% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.