ClawHub
Back to skill

Security audit

SOX Compliance

Security checks for vulnerabilities and agentic risk

Overview

This SOX assessment skill is purpose-aligned, but it sends potentially sensitive compliance details to an external API without enough data-handling disclosure.

Review this skill before installing in a real compliance environment. Use it only with organization-approved data sharing, avoid putting raw audit evidence, secrets, customer or employee data, financial details, or confidential control weaknesses in notes, and confirm the provider's retention, access, deletion, residency, and contractual controls before sending SOX assessment content.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

other

Warning
Location
SKILL.md:31
Finding

Potential Disclosure of Sensitive Compliance Data to an External Service

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:31-57, SKILL.md:100-126, and SKILL.md:185-188
Vulnerability Type: Sensitive Data Exposure
Risk Level: Medium

Vulnerable Documentation

The sample request encourages users to include internal audit findings and supporting evidence in the request:

json
{
  "tier": "standard",
  "sessionId": "sess_20250117_audit_001",
  "controls": {
    "financial_reporting": [
      {
        "controlId": "FR-001",
        "compliant": true,
        "notes": "Monthly reconciliation process documented and tested"
      },
      {
        "controlId": "FR-002",
        "compliant": false,
        "notes": "Missing evidence of Q3 review sign-off"
      }
    ],
    "it_general_controls": [
      {
        "controlId": "IT-101",
        "compliant": true,
        "notes": "Access reviews performed quarterly"
      },
      {
        "controlId": "IT-102",
        "compliant": true
      }
    ]
  }
}

The documented endpoint accepts those values:

text
### POST /api/sox/assess

**Description:** Perform comprehensive SOX compliance assessment across multiple control domains and sections.

**Method:** POST

**Path:** `/api/sox/assess`

The documentation identifies an externally operated API route:

text
## References

- **Kong Route:** https://api.mkkpro.com/compliance/sox-compliance
- **API Docs:** https://api.mkkpro.com:8039/docs

Technical Analysis

The skill is designed to process organizational SOX assessment information, including control identifiers, compliance decisions, session identifiers, and free-form notes. The example specifically demonstrates transmitting details about a missing review sign-off, which may reveal an internal control deficiency.

The referenced service is hosted at api.mkkpro.com, outside the local skill package. However, the documentation provides no warn ...[truncated 2144 chars]

Remediation
View remediation

Remediation Suggestions

  1. Clearly disclose that assessment content is transmitted to and processed by an externally operated service.
  2. Require explicit user or organizational approval before sending compliance data outside the local environment.
  3. Add prominent instructions prohibiting secrets, credentials, personal data, raw audit evidence, and unnecessary financial details in notes.
  4. Apply client-side redaction or data-loss-prevention checks before submission.
  5. Constrain the length and permitted content of sessionId, controlId, section names, and notes.
  6. Document the service's data retention, deletion, encryption, access-control, residency, incident-response, and secondary-use policies.
  7. Provide a self-hosted or organization-approved endpoint option for sensitive assessments.
  8. Minimize transmitted data by using standardized finding codes instead of free-form evidence wherever possible.
  9. Establish contractual safeguards and verify that third-party handling is compatible with the organization's SOX, privacy, and vendor-risk requirements.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs users to send potentially sensitive SOX assessment data, including session identifiers and free-form compliance notes, to a remote API without any warning about data sensitivity, retention, jurisdiction, or third-party handling. In a compliance context, those notes may contain audit evidence, control gaps, or internal process details that could create confidentiality, regulatory, and reputational risk if transmitted inappropriately.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
77% confidence
Finding

The skill references an external API endpoint, which confirms data leaves the local environment and is sent to infrastructure outside the immediate agent context. While external APIs are not inherently malicious, this becomes security-relevant here because the skill is designed to process SOX compliance information, making outbound transmission of audit-related content more sensitive than in a low-risk domain.

Content

Scanner excerpt · SKILL.md (reported line 187)May include surrounding context.

md
## References

- **Kong Route:** https://api.mkkpro.com/compliance/sox-compliance
- **API Docs:** https://api.mkkpro.com:8039/docs

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

This OpenAPI manifest describes a POST endpoint that performs a compliance assessment and requires submission of controls and a session identifier, but it provides no warning that user-supplied assessment data will be transmitted to a remote service. For manifest files, missing-warning concerns apply here because the description omits any disclosure about privacy or data-handling implications of sending enterprise compliance information.

Content

No source excerpt is available for this finding.