Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cross-framework compliance readiness orchestrator. Use when preparing for multi-framework certification (SOC 2 + ISO 27001 + NIST CSF), building a shared-evidence strategy, sequencing certifications, or mapping a control across frameworks.
.claude/skills/borghei-compliance-readiness/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-20 | ✓→✓ | = Same ✓ | 148% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 90% | 0% |
| case-21 | ✓→✓ | = Same ✓ | 240% | 0% |
The orchestrator skill for organizations pursuing multiple compliance frameworks. Reduces duplication, accelerates certification, and shares evidence across SOC 2, ISO 27001, NIST CSF, GDPR, HIPAA, and others.
When to use this skill vs. framework-specific audit-prep:
soc2-audit-prep, gdpr-audit-prep, etc.): single-framework sprint| Situation | Skill applies | |-----------|---------------| | Pursuing SOC 2 + ISO 27001 + NIST CSF in parallel | Yes — start here | | Healthcare org pursuing SOC 2 + HIPAA + ISO 27001 | Yes | | Building shared-evidence platform | Yes — see shared evidence strategy | | Mapping one control to multiple frameworks | Yes — scripts/shared_evidence_finder.py | | Deciding which framework to certify first | Yes — see sequencing decisions | | Generating multi-framework roadmap | Yes — scripts/readiness_roadmap_generator.py | | Single-framework audit prep | Use framework-specific skill |
Most controls are shared across compliance frameworks. A well-designed control catalog satisfies multiple frameworks simultaneously. Without coordination, you build separate evidence + procedures per framework — 3x the work, 3x the maintenance, 3x the auditor confusion.
Common shared controls:
| Control area | SOC 2 | ISO 27001 | NIST CSF | NIS2 | DORA | PCI-DSS | HIPAA | GDPR | |--------------|-------|-----------|----------|------|------|---------|-------|------| | Access control | CC6.1 | A.8.5 | PR.AA | Art.21.2.j | Art.9.4 | Req 7-8 | §164.312(d) | Art.32 | | Encryption | CC6.7 | A.8.24 | PR.DS | Art.21.2.h | Art.9.2 | Req 3-4 | §164.312(a)(2)(iv) | Art.32 | | Incident response | CC7.4 | A.5.24 | RS.MA | Art.23 | Art.17 | Req 12.10 | §164.308(a)(6) | Art.33 | | Risk assessment | CC3.1 | Cl.6.1 | ID.RA | Art.21.1 | Art.6 | Req 12.2 | §164.308(a)(1) | Art.35 | | Logging | CC7.2 | A.8.15 | DE.CM | Art.21.2.b | Art.10 | Req 10 | §164.312(b) | Art.30 | | Vendor management | CC9.2 | A.5.19 | GV.SC | Art.21.2.d | Art.28 | Req 12.8 | §164.308(b) | Art.28 |
See references/control-mapping-soc2-iso27001-nist.md for the full mapping.
Which framework to pursue first?
SaaS / Tech (B2B enterprise customers):
Healthcare (US):
FinTech / financial services (EU):
Medical devices:
Build one control catalog covering all in-scope frameworks. Each control maps to multiple frameworks. One implementation; one evidence trail.
Control: Access Reviews (Quarterly)
- SOC 2: CC6.3
- ISO 27001: A.5.18
- NIST CSF: PR.AA-04
- HIPAA: §164.308(a)(4)
- GDPR: Art.32 (security of processing)
Evidence: Quarterly access-review records, signed by team lead
Frequency: Quarterly
Owner: IT SecurityOne artifact satisfies five frameworks.
Single source-of-truth for evidence (Drata / Vanta / Thoropass / Sprinto / homegrown):
Annual management review covers all frameworks:
Plan internal audit to cover overlapping clauses:
Weeks 1-4: Common control catalog build; gap analysis per framework
Weeks 5-8: Gap remediation (technical + procedural)
Weeks 9-12: Evidence collection + walkthroughs
Weeks 13-14: SOC 2 audit
Weeks 15-16: ISO 27001 Stage 1
(then ISO 27001 Stage 2 ~4-8 weeks later)Weeks 1-2: Audit readiness assessment per framework
Weeks 3-6: Gap remediation
Weeks 7-9: Walkthroughs + evidence finalization
Weeks 10-12: Audits (sequential or parallel depending on auditor capacity)Before generating the roadmap, 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 roadmap.
python3 scripts/multi_framework_scorer.py --config controls.yamlpython3 scripts/shared_evidence_finder.py --frameworks SOC2,ISO27001python3 scripts/readiness_roadmap_generator.py --target-frameworks SOC2,ISO27001,GDPR| Script | Purpose | |--------|---------| | scripts/multi_framework_scorer.py | Score readiness across multiple frameworks | | scripts/shared_evidence_finder.py | Identify shared controls; map evidence to frameworks | | scripts/readiness_roadmap_generator.py | Generate multi-framework readiness roadmap |
ra-qm-team/soc2-compliance-expert — deep SOC 2 programra-qm-team/information-security-manager-iso27001 — deep ISO 27001 programra-qm-team/nist-csf-specialist — deep NIST CSF programra-qm-team/gdpr-dsgvo-expert — deep GDPR programra-qm-team/fda-consultant-specialist — deep FDA programra-qm-team/infrastructure-compliance-auditor — cross-framework infra auditra-qm-team/audit-prep/* — framework-specific audit-prep skills| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 29,934 | 26,776 | -11% | 1 | 1 | 0% | 4,820 | 7,069 | +47% | 0 | 0 | — |
case-02 | fail→fail | 25,258 | 28,400 | +12% | 1 | 1 | 0% | 4,261 | 7,343 | +72% | 0 | 0 | — |
case-20 | pass→pass | 11,672 | 11,384 | -2% | 1 | 1 | 0% | 1,777 | 4,408 | +148% | 0 | 0 | — |
case-03 | fail→fail | 25,298 | 28,015 | +11% | 1 | 1 | 0% | 3,988 | 7,316 | +83% | 0 | 0 | — |
case-04 | fail→pass | 20,714 | 9,772 | -53% | 1 | 1 | 0% | 3,123 | 4,138 | +33% | 0 | 0 | — |
case-05 | pass→pass | 13,248 | 8,458 | -36% | 1 | 1 | 0% | 2,068 | 3,930 | +90% | 0 | 0 | — |
case-06 | fail→pass | 19,372 | 16,880 | -13% | 1 | 1 | 0% | 3,269 | 5,454 | +67% | 0 | 0 | — |
case-21 | pass→pass | 8,043 | 10,835 | +35% | 1 | 1 | 0% | 1,264 | 4,294 | +240% | 0 | 0 | — |
case-07 | pass→pass | 14,709 | 13,044 | -11% | 1 | 1 | 0% | 2,308 | 4,703 | +104% | 0 | 0 | — |
case-08 | pass→pass | 9,841 | 13,162 | +34% | 1 | 1 | 0% | 1,572 | 4,710 | +200% | 0 | 0 | — |
case-09 | pass→pass | 12,563 | 17,607 | +40% | 1 | 1 | 0% | 2,084 | 5,489 | +163% | 0 | 0 | — |
case-10 | fail→fail | 14,748 | 15,605 | +6% | 1 | 1 | 0% | 2,740 | 5,473 | +100% | 0 | 0 | — |
case-11 | pass→pass | 12,894 | 15,970 | +24% | 1 | 1 | 0% | 2,427 | 5,391 | +122% | 0 | 0 | — |
case-12 | pass→pass | 15,660 | 12,672 | -19% | 1 | 1 | 0% | 2,658 | 4,802 | +81% | 0 | 0 | — |
case-13 | pass→pass | 17,204 | 14,633 | -15% | 1 | 1 | 0% | 3,129 | 5,455 | +74% | 0 | 0 | — |
case-14 | pass→pass | 16,878 | 14,168 | -16% | 1 | 1 | 0% | 3,015 | 5,167 | +71% | 0 | 0 | — |
case-15 | pass→pass | 5,677 | 6,584 | +16% | 1 | 1 | 0% | 1,012 | 3,763 | +272% | 0 | 0 | — |
case-16 | pass→pass | 5,337 | 7,730 | +45% | 1 | 1 | 0% | 939 | 3,903 | +316% | 0 | 0 | — |
case-17 | fail→fail | 12,446 | 13,201 | +6% | 1 | 1 | 0% | 1,922 | 4,668 | +143% | 0 | 0 | — |
case-18 | pass→pass | 10,214 | 11,044 | +8% | 1 | 1 | 0% | 1,633 | 4,301 | +163% | 0 | 0 | — |
case-19 | pass→pass | 12,225 | 10,521 | -14% | 1 | 1 | 0% | 1,984 | 4,272 | +115% | 0 | 0 | — |
case-22 | pass→pass | 12,885 | 13,447 | +4% | 1 | 1 | 0% | 2,112 | 4,868 | +130% | 0 | 0 | — |
case-23 | pass→pass | 10,558 | 8,935 | -15% | 1 | 1 | 0% | 1,795 | 3,992 | +122% | 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. 23 cases were attempted. The headline lift of +9 percentage points is the difference between those two pass rates over the 23 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.