ClawHub
Back to skill

Security audit

Paste Rs

Security checks across malware telemetry and agentic risk

Overview

This skill publicly uploads user-selected text to paste.rs and saves a local copy first, but those behaviors are disclosed and fit its purpose.

Install only if you are comfortable sending selected text or files to a public paste service. Keep redaction enabled, review content before upload, avoid secrets or proprietary material, and delete the saved local .md file if you do not want a retained copy.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill invokes a Python script and curl to read local input, write a local .md file, and send content to an external service, but it does not declare corresponding permissions. Undeclared file and network capabilities reduce transparency and can cause sensitive data to be exfiltrated or persisted locally without appropriate review, especially because the skill is designed to upload arbitrary user-provided text.

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill persistently writes the full pasted content to a local file in /tmp before uploading, even though its advertised behavior is just to send text to paste.rs and return a URL. This creates an additional data exposure surface: sensitive logs, configs, or tokens may remain on disk after execution, be readable by other local users/processes depending on environment, or be recovered later from temporary storage.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.