ClawHub
Back to skill

Security audit

Xpcomment2

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local image-metadata helper, but users should be careful because its write and bulk examples can change files in place.

Install this only if you want an agent to manage XPComment metadata on local images. Before allowing add, remove, wildcard, or recursive actions, confirm the exact files or directory scope and keep backups when the metadata matters.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill defines broad trigger conditions and says it must be used immediately for generic image-comment requests, which can cause the agent to invoke file-modifying behavior too aggressively. In a skill that can overwrite or delete metadata, overbroad routing increases the chance of unintended file changes or bypassing safer clarification steps.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documented add and remove commands perform in-place metadata modification with -overwrite_original, and the remove action deletes the field, but the skill does not warn users that these changes are destructive and may be hard to recover. This is dangerous because an agent could execute a write/delete operation on the wrong file or with the wrong text without prompting the user to confirm.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The bulk-processing examples include wildcard and recursive commands that can modify many images in place, including all matching files under a directory tree, without any warning about scale or irreversibility. In agentic use, this raises the risk of mass unintended metadata changes across large collections if the path or pattern is wrong.

Static analysis

No suspicious patterns detected.