Back to skill

Security audit

image-face-recognition

Security checks for vulnerabilities and agentic risk

Overview

This face-recognition skill is small and not overtly malicious, but it is under-scoped for sensitive biometric image processing.

Only install this for explicit, consented face-recognition workflows. Treat uploaded images as biometric personal data, verify what external recognition service the missing script would call, and confirm image retention, consent, and API-key handling before use.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documentation describes face recognition and an API key requirement but does not warn that this involves sensitive biometric processing and possible transmission of images to an external service. Users or orchestrators may enable or invoke the skill without understanding the privacy, consent, retention, and compliance implications of processing face data, increasing the risk of inappropriate handling of highly sensitive personal information.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill's activation guidance is extremely broad ('ai related functionality', 'recognition tasks', 'image operations'), which can cause the agent to invoke this face-recognition skill for unrelated image requests. Because face recognition is a sensitive biometric capability, overbroad routing increases the chance of unnecessary processing of user images and inadvertent exposure of personal data to the skill or its backend API.

Content

No source excerpt is available for this finding.