ClawHub

oc-doctor

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate local OpenClaw diagnostic skill, but its fixes can change or delete local OpenClaw files if you approve them.

Install only if you want a tool to inspect your local OpenClaw configuration, sessions, logs, cron data, and workspace markdown. Review the diagnostic report before approving fixes, prefer one-by-one changes for config/security/session repairs, and avoid rendering the demo SVG in privacy-sensitive offline environments if third-party font loading matters.

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

Low
Confidence
94% confidence
Finding
The SVG pulls a font from fonts.googleapis.com, which creates an unnecessary external network dependency in a skill described as a local diagnostic tool. Even though this is only a demo asset, rendering the SVG can leak user metadata such as IP address and user agent to a third party and can fail in offline or restricted environments.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list includes broad natural-language phrases such as "openclaw doctor" and "openclaw diagnose", which can overlap with ordinary user conversation or troubleshooting requests. In an agent skill system, this can cause unintended activation of a powerful diagnostic/fix workflow, potentially exposing local configuration data or prompting repair actions when the user did not explicitly intend to invoke the skill.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The document promotes bulk automatic remediation and destructive actions such as cache cleanup and file replacement/deletion without clearly warning about possible data loss, side effects, or the need for confirmation and backup. In a troubleshooting skill that offers one-click fixes, this can normalize unsafe operation and increase the chance a user approves impactful changes they do not fully understand.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal