ClawHub
Back to skill

Security audit

Pptx Pdf Font Fix Tool Free

Security checks across malware telemetry and agentic risk

Overview

This skill is mostly a local PPT font-repair helper, but it asks for broad read/write/exec authority and includes an under-explained callback URL feature.

Review this skill before installing. Use it only on PPT files you intend to process, avoid providing a callback URL unless you understand where results may be sent, and expect the agent may run local commands or write repaired/PDF outputs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The introduction of a callback_url and network-oriented behavior expands the skill's trust boundary beyond local file processing into outbound communication. In a tool presented as a local PPT/font repair utility, this can enable silent data exfiltration, SSRF-like requests, or disclosure of file metadata/results to untrusted endpoints if an agent honors the parameter.

Vague Triggers

Medium
Confidence
78% confidence
Finding
Overly broad trigger phrases increase the chance that the skill activates during ordinary conversation and performs file-processing or execution-adjacent actions without sufficiently specific user intent. In an agent environment with read/write/exec tools, ambiguous invocation materially raises the risk of unintended file access or command execution.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill declares file processing, PDF conversion, and exec-capable behavior without clearly warning users that it may read local files, write outputs, and invoke local tools. In a privileged agent context, lack of transparent disclosure undermines informed consent and can lead to unintended exposure or modification of user data.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.