Back to skill

Security audit

Insurance Anti Fraud

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign because it found no reported persistence or purpose-capability concerns, while flagging privacy safeguards for sensitive external checks; the review is incomplete and confidence is medium.

Before installing, consider whether the skill's suggested external checks fit your work. Confirm that you have authority to share claimant information, use approved services, and limit what you disclose. The verdict is benign, but the scan was partial: one of two components was scanned, the manifest could not be parsed, and several references were missing, so the evidence is incomplete.

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
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

The investigation workflow calls for sensitive external data checks without warning users about the associated disclosure of claimant information.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The investigation workflow recommends obtaining medical, insurance, and third-party risk data without requiring verification of authorization, a lawful basis, approved recipients, or minimum necessary disclosure. If operationalized by a tool-enabled agent, these checks could expose claimant identifiers or sensitive health information to unauthorized services. Such checks are legitimate in insurance investigations, and the later confidentiality requirement reduces concern, but it does not establish safeguards for external data sharing; no actual transmission or malicious destination is shown.

Content

No source excerpt is available for this finding.