Skip to content

[REVIEW] hipaa-review: add source-dated regulatory and threat-evidence gates #179

@hermesbountyhunter

Description

@hermesbountyhunter

Skill Being Reviewed

Skill name: hipaa-review
Skill path: skills/compliance/hipaa-review/

False Positive Analysis

Benign evidence that can be over-weighted when the review lacks source-date gates:

Risk analysis includes a healthcare destructive-malware scenario.
Backup plan says nightly backups are retained.
Training deck mentions ransomware and wiper malware.
BAA inventory exists.

Why this is a false positive:
Those artifacts can be useful, but they do not prove HIPAA Security Rule readiness unless the reviewer records (1) which regulatory or OCR source is being relied on, (2) the source date/version, (3) whether the threat example is documented rather than anecdotal, and (4) whether the backup/recovery evidence was actually tested. The current skill includes very specific current-threat statements and monetary penalty figures inline, but the output does not require citations, source dates, or a confidence level for those time-sensitive claims. A report can therefore look more current and authoritative than the evidence supports.

Coverage Gaps

Missed variant 1: Time-sensitive OCR/enforcement and penalty claims are not source-dated

Finding: "High enforcement risk; potential civil monetary penalties ($100-$50,000 per violation, annual max $2,067,813 per identical violation category per calendar year as of 2024 penalty tiers)"
Evidence source: not recorded
Source date: not recorded
Inflation-adjusted tier year: not recorded

Why it should be caught:
HIPAA civil monetary penalty caps and OCR enforcement priorities can change. The skill should force the reviewer to record the exact HHS/OCR or Federal Register source and effective year before putting penalty amounts or "#1 most cited" style claims into a client-facing assessment. Without that, stale figures can be repeated as current compliance guidance.

Missed variant 2: Threat-intel examples are embedded without evidence provenance

Risk analysis notes a named destructive/wiper attack against a healthcare supplier.
Citation: missing
Date observed: missing
Actor attribution confidence: missing
Applicability to covered entity / business associate: not mapped

Why it should be caught:
Threat scenarios are valuable for contingency planning, but named incidents and actor attribution are high-risk facts. The review should require a threat-evidence table with source, publication date, confidence, affected sector, and the specific HIPAA safeguard it informs. Otherwise a speculative or stale incident can drive findings that should instead be framed as generic destructive-malware preparedness.

Missed variant 3: Backup recoverability is asserted without restoration evidence

Control: immutable backups configured
Evidence: backup policy and console screenshot
Missing: restore test date, sample system restored, RPO/RTO result, privileged-access failure mode, backup deletion protection test

Why it should be caught:
164.308(a)(7)(ii)(A)-(E) and 164.312(c)(1)/(c)(2) reviews should distinguish having backup mechanisms from proving ePHI can be restored with integrity after destructive malware or administrator compromise. The skill mentions immutable/offline backups, but the output does not require a recovery-test evidence row.

Missed variant 4: BAA/subcontractor evidence is not tied to source freshness and downstream obligations

Vendor: transcription SaaS
BAA: yes
Subcontractors: not reviewed
Security incident reporting window: not extracted
Termination/return-destruction clause: not extracted
Last review date: missing

Why it should be caught:
For 164.308(b)(1) and 164.314(a), a BAA's existence is not enough. The reviewer should capture contract date, covered services, subcontractor flow-down, breach/security-incident notice obligations, termination/return-destruction language, and review freshness.

Edge Cases

  • A healthcare organization may have strong ransomware controls but weak ePHI integrity verification after restore; the review should not mark contingency planning complete without restore evidence.
  • A sourced threat report may describe a non-HIPAA entity or uncertain attribution; the review should let the assessor use it as contextual risk input without turning it into a definitive compliance finding.
  • Addressable encryption decisions may be valid alternatives, but only if the documented rationale is current and tied to actual ePHI flows rather than generic vendor assurances.

Remediation Quality

  • Fix resolves the vulnerability
  • Fix doesn't introduce new security issues
  • Fix doesn't break functionality
  • Issues found: Add explicit source/provenance fields to the review output before using time-sensitive penalty, enforcement, or threat-intelligence statements in final findings. Add a recovery-test evidence row for contingency plan and ePHI integrity conclusions.

Comparison to Other Tools

Tool Catches this? Notes
Semgrep No Semgrep can flag code/config patterns but will not validate HIPAA regulatory-source freshness or BAA clause completeness.
CodeQL No CodeQL is useful for application vulnerabilities, not compliance evidence provenance.
HHS OCR Audit Protocol / official guidance Partial Provides the authoritative source material, but the skill needs to require source/date capture and mapping into the report.
GRC platforms (Drata/Vanta/Hyperproof-style evidence workflows) Partial They often track evidence owners and freshness; the skill should emulate that evidence discipline for generated assessments.

Overall Assessment

Strengths:

  • Good structure across Administrative, Physical, Technical, Organizational, and Documentation requirements.
  • Correctly explains that addressable specifications are not optional.
  • Useful emphasis on destructive-malware readiness and backup resilience.

Needs improvement:

  • Time-sensitive regulatory, penalty, and threat-intelligence claims need source/date/confidence fields.
  • Contingency planning needs restore-test evidence, not only backup-policy evidence.
  • BAAs need clause-level and subcontractor-flow-down evidence rather than a simple present/missing inventory.

Priority recommendations:

  1. Add a Regulatory and threat source register section: source name, URL/reference, publication/effective date, reviewed date, claim supported, and confidence.
  2. Add a Contingency and restore evidence table: system, ePHI data class, backup type, immutability/deletion protection, last restore test, RPO/RTO result, integrity verification, and exception owner.
  3. Add BAA/subcontractor evidence fields: covered service, contract date, incident notice terms, subcontractor flow-down, termination/return-destruction terms, and last review date.

Bounty Info

  • I have read and agree to the CONTRIBUTING.md bounty terms
  • Preferred payment method: PayPal (can provide details privately if accepted)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions