Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Singapore - Personal Data Protection Ac (PDPA) (2012) expert. Reference-depth framework plugin with assessment, scope determination, and evidence checklist — backed by the SCF crosswalk. Level up to Full by adding framework-specific workflow commands.
.claude/skills/grcengclub-singapore-pdpa-expert/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -36% | 0% |
Reference-depth expertise for Singapore - Personal Data Protection Ac (PDPA) (2012). This plugin bundles the SCF crosswalk (30 SCF controls → 14 framework controls) with framework-specific context.
apac-sgp-pdpa-2012TODO: add the following when filling in expertise. Reference depth expects all of these to have at least two-sentence answers.
TODO: one-paragraph summary of what Singapore - Personal Data Protection Ac (PDPA) (2012) exists to do and who it affects. Avoid verbatim regulation text — paraphrase and cite articles/sections by number.
TODO: who must comply, where operations must be happening, what the carve-outs are.
TODO: the named documents or registers the framework requires (e.g. GDPR's ROPA/DPIA, PCI DSS's ROC/SAQ, ISO 27001's Statement of Applicability). Stub if uncertain; a Full-depth PR can add commands to generate each.
TODO: notification windows (e.g. GDPR 72 hours), assessment frequency, recertification cycles.
TODO: who enforces, how penalties are calculated, recent enforcement patterns worth knowing.
TODO: overlaps with GDPR / ISO / NIST / sectoral rules. Reference the cross-framework analyzer outputs when useful.
TODO: at least 2–3 community-level misunderstandings this plugin should correct. Example: "ITAR only affects munitions" (it covers technical data too).
/singapore-pdpa:scope — determine applicability/singapore-pdpa:assess — run a gap assessment/singapore-pdpa:evidence-checklist — enumerate evidence requirementsAll three delegate to /grc-engineer:gap-assessment with SCF framework ID apac-sgp-pdpa-2012 for the control-by-control mechanics, and wrap the results in Singapore - Personal Data Protection Ac (PDPA) (2012)-specific terminology.
Full-depth plugins add framework-specific workflow commands tied to the audit ritual. Candidates for this framework:
TODO: propose 2–4 framework-specific commands. Examples from existing Full-depth plugins:
soc2: /soc2:service-auditor-prep, /soc2:trust-service-matrixfedramp-rev5: /fedramp-rev5:poam-review, /fedramp-rev5:ssp-outlinepci-dss: /pci-dss:roc-walkthrough, /pci-dss:saq-routecmmc: /cmmc:c3pao-readiness| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,165 | 18,265 | +13% | 1 | 1 | 0% | 2,565 | 3,695 | +44% | 0 | 0 | — |
case-02 | fail→fail | 9,188 | 11,866 | +29% | 1 | 1 | 0% | 1,375 | 2,702 | +97% | 0 | 0 | — |
case-03 | fail→fail | 18,906 | 16,628 | -12% | 1 | 1 | 0% | 2,860 | 3,357 | +17% | 0 | 0 | — |
case-04 | fail→pass | 17,959 | 13,194 | -27% | 1 | 1 | 0% | 2,802 | 2,802 | 0% | 0 | 0 | — |
case-05 | pass→pass | 20,040 | 15,945 | -20% | 1 | 1 | 0% | 3,255 | 3,204 | -2% | 0 | 0 | — |
case-10 | fail→pass | 12,314 | 2,585 | -79% | 1 | 1 | 0% | 1,793 | 1,189 | -34% | 0 | 0 | — |
case-06 | fail→pass | 10,687 | 10,785 | +1% | 1 | 1 | 0% | 1,688 | 2,567 | +52% | 0 | 0 | — |
case-07 | fail→pass | 6,544 | 2,352 | -64% | 1 | 1 | 0% | 889 | 1,156 | +30% | 0 | 0 | — |
case-08 | fail→pass | 20,636 | 2,261 | -89% | 1 | 1 | 0% | 1,711 | 1,102 | -36% | 0 | 0 | — |
case-09 | fail→pass | 8,178 | 1,649 | -80% | 1 | 1 | 0% | 1,122 | 997 | -11% | 0 | 0 | — |
case-11 | fail→pass | 12,341 | 2,564 | -79% | 1 | 1 | 0% | 1,786 | 1,169 | -35% | 0 | 0 | — |
case-12 | pass→pass | 6,738 | 1,972 | -71% | 1 | 1 | 0% | 900 | 1,071 | +19% | 0 | 0 | — |
case-13 | pass→pass | 2,349 | 1,339 | -43% | 1 | 1 | 0% | 324 | 906 | +180% | 0 | 0 | — |
case-14 | fail→pass | 12,571 | 2,662 | -79% | 1 | 1 | 0% | 1,731 | 1,164 | -33% | 0 | 0 | — |
case-15 | fail→pass | 10,987 | 2,052 | -81% | 1 | 1 | 0% | 1,633 | 1,077 | -34% | 0 | 0 | — |
case-16 | fail→pass | 20,435 | 18,265 | -11% | 1 | 1 | 0% | 3,101 | 3,555 | +15% | 0 | 0 | — |
case-17 | fail→pass | 8,411 | 3,556 | -58% | 1 | 1 | 0% | 1,224 | 1,063 | -13% | 0 | 0 | — |
case-18 | pass→pass | 13,977 | 10,406 | -26% | 1 | 1 | 0% | 2,116 | 2,292 | +8% | 0 | 0 | — |
case-19 | pass→pass | 18,378 | 11,348 | -38% | 1 | 1 | 0% | 2,727 | 2,481 | -9% | 0 | 0 | — |
case-20 | fail→pass | 14,900 | 1,942 | -87% | 1 | 1 | 0% | 2,226 | 999 | -55% | 0 | 0 | — |
case-21 | fail→pass | 14,187 | 1,930 | -86% | 1 | 1 | 0% | 2,023 | 1,034 | -49% | 0 | 0 | — |
case-22 | fail→pass | 14,272 | 8,494 | -40% | 1 | 1 | 0% | 2,238 | 2,127 | -5% | 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 +64 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.