Back to skill

Security audit

face-shape-analysis

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because it reports that personal facial measurements may be sent to an external service without a privacy warning or informed permission.

Before installing, consider whether you are comfortable with facial measurements being sent to the named external service. The scanner recommends disclosing what will be sent and obtaining permission before sharing personal measurements. The scan covered the single reported component completely, but this verdict is suspicious because of the privacy concern.

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 workflow sends personal facial measurements to an external service without a privacy warning.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The workflow directs the agent to transmit a person's facial measurements to an external MCP service without first disclosing that transfer or obtaining informed permission. These measurements are personal physical characteristics that could be retained or associated with request metadata, although four distances alone do not establish a uniquely identifying biometric profile. The explicitly named service and task-relevant transmission reduce suspicion of malicious exfiltration, but the service's read-only status does not prevent collection or logging of submitted data.

Content

No source excerpt is available for this finding.