ClawHub
Back to skill

Security audit

Docker Socket Proxy

Security checks across malware telemetry and agentic risk

Overview

This is a powerful Docker administration skill, but its Docker control and data-reading capabilities are disclosed, purpose-aligned, and scoped through the user's docker-socket-proxy configuration.

Install only if you intend to let an agent administer Docker through a tightly scoped docker-socket-proxy. Enable the minimum proxy sections needed, avoid exposing a raw Docker daemon, and require explicit human approval before exec, restart, kill, stop, or prune commands, especially on production hosts.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly exposes destructive Docker operations such as stop, kill, restart, prune, and rename against a remote Docker host, but the description does not clearly warn that these actions can disrupt production workloads or permanently remove resources. In context, Docker API access can directly affect availability and data retention, so presenting these actions without prominent impact warnings increases the chance of unsafe or unintended use.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill advertises access to logs, stats, inspection data, and Swarm secrets/config-related endpoints without clearly warning that these outputs may contain sensitive information such as credentials, tokens, environment variables, internal topology, or operational metadata. Because the target is a remote Docker host, the context makes this more dangerous: a user may unintentionally expose sensitive infrastructure data through routine diagnostic actions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.