ClawHub
Back to skill

Security audit

screenshot.fyi

Security checks across malware telemetry and agentic risk

Overview

The skill appears to do what it claims overall, but it mislabels screenshot creation as a safe read-only action and may run credentialed, billable URL captures without explicit confirmation.

Install only if you are comfortable letting the agent use your connected OOMOL screenshot.fyi account to capture URLs. Before running it, confirm the exact URL and expected screenshot action, especially for private, internal, sensitive, or billable targets.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (2)

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The skill metadata and description claim the skill is for 'searching and reading data', but the only documented action is `take_screenshot`, which creates a new artifact and can trigger external network activity against arbitrary URLs. This mismatch can mislead an agent into treating the action as read-only, bypassing confirmation or safety checks that should apply to state-changing operations.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The safety section states that untagged actions are reads, yet `take_screenshot` is untagged even though it creates a screenshot and consumes service-side resources. An agent following this guidance may execute screenshot creation without confirmation, causing unintended external requests, billing impact, or interaction with sensitive/internal URLs if user input is not constrained.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.