ClawHub
Back to skill

Security audit

iwork2md

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent local converter for Apple iWork files, with expected file access and no evidence of hidden network, credential, persistence, or destructive behavior.

Install this only if you want the agent to read specified iWork documents and convert their contents to Markdown. Be explicit about which file to process, and use stdout or an explicit output path when handling sensitive documents.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill clearly instructs the user/agent to run local Python code that reads arbitrary iWork files and can write Markdown output next to the source or to an explicit path, but the skill metadata does not declare those file access capabilities. This is a real security-relevant issue because undeclared read/write behavior reduces transparency and can bypass policy controls or user expectations about what the skill is allowed to do.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The README advertises very broad natural-language trigger phrases like 'read that Keynote file' and 'extract text from my Numbers sheet', which can match ordinary conversation and cause the skill to invoke when the user did not explicitly intend file conversion. For a skill that processes local files, unintended invocation can expose document contents or cause the agent to operate on sensitive files based on ambiguous references.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.