Back to skill

Security audit

ai-insurance-advisor

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: most reviewed security areas were rated OK, while one finding flags a risk that insurance coverage conditions may be presented as unconditional benefits.

The verdict is benign, not risk-free, and confidence is medium. Consider that coverage claims may omit conditions that affect reimbursement. The scan covered all seven listed components, but analysis was partial because some local references could not be resolved; evidence is therefore 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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Conditional structured coverage decisions can become unconditional treatment-benefit claims.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

_structured_value_state can classify an entry as affirmative when covered is true without validating or preserving accompanying eligibility conditions, such as restrictions to designated hospitals or approved indications. That state can resolve conflicting coverage evidence, after which _medical_treatment_suffix emits an unconditional treatment-benefit claim; the text-based uncertainty checks do not reliably capture structured or unrecognized conditions. In an insurance-advice skill, this can mislead users into selecting coverage that will not reimburse their treatment, although the supplied code does not purchase policies or transmit data.

Content

No source excerpt is available for this finding.