ClawHub

claim-risk-auditor

Security checks across malware telemetry and agentic risk

Overview

This skill has a reasonable text-auditing purpose, but it can read and print the full system clipboard under broad prompts without a clear confirmation step.

Install only if you are comfortable with the agent reading and displaying your current clipboard when this skill runs. Clear sensitive clipboard contents first, prefer pasting the text directly when possible, and watch for future versions that add explicit confirmation before clipboard access.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The script accesses the system clipboard via `pbpaste` and prints its full contents, which can expose unrelated sensitive local data such as passwords, API keys, private messages, or proprietary text. For a skill whose stated purpose is auditing provided text, silently pulling arbitrary clipboard contents is broader than necessary and increases the risk of unintended data collection or disclosure.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The invocation phrases are broad enough to match common editing or review requests, which can cause the skill to trigger unexpectedly in situations where the user did not explicitly ask to use this particular tool. Because the skill then reads clipboard contents, overbroad activation increases the chance of unintended access to sensitive data and accidental processing of unrelated material.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill instructs the agent to read from the clipboard via pbpaste without any user-facing disclosure or confirmation at the point of use. Clipboard contents often contain passwords, API keys, personal data, or proprietary text, so silent access can expose sensitive information beyond what the user intended to share.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal