ClawHub
Back to skill

Security audit

PaperPod

Security checks across malware telemetry and agentic risk

Overview

PaperPod is a documentation-only skill for a remote sandbox service, and its powerful execution, networking, and persistence features are disclosed and aligned with that purpose.

Install only if you want your agent to use PaperPod's external sandbox. Treat the PaperPod token like a password, avoid sending secrets or private files unless necessary, review the separate @paperpod/cli package before installing it globally, do not expose sensitive or unauthenticated services publicly, and clean up background processes, exposed ports, browser sessions, and persistent memory after use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documentation instructs users to place long-lived bearer tokens directly on the command line (`ppod login pp_sess_...` and `Authorization: Bearer pp_sess_...`). Tokens passed this way can be exposed through shell history, terminal logs, screenshots, CI logs, or process listings on multi-user systems, increasing the chance of credential theft.

Missing User Warnings

Low
Confidence
90% confidence
Finding
This reference document prominently demonstrates shell execution, network access, package installation, Git cloning, and subprocess usage without any warning about side effects such as outbound data transfer, remote code retrieval, filesystem modification, or command injection risks when interpolating untrusted input. In an isolated execution product, these capabilities are expected, but omitting safety guidance can still lead users or downstream agents to perform risky actions unsafely.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.