ClawHub
Back to skill

Security audit

Clawdbot Logs

Security checks across malware telemetry and agentic risk

Overview

This is a diagnostic skill that reads OpenClaw/Clawdbot logs and session files locally; its sensitive log access is expected for troubleshooting, but users should handle outputs carefully.

Install only if you are comfortable letting the agent inspect local OpenClaw/Clawdbot logs and session JSONL files. Treat outputs as private: review and redact conversation text, identifiers, token usage, costs, and tool details before sharing them with anyone.

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 (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The markdown explicitly states that session files contain full conversations, token usage, and costs, and later provides a command to extract recent user message text. The skill does not include any caution that these diagnostics may reveal sensitive user data or should be handled carefully.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
This markdown file offers commands to inspect journal entries and diagnostic logs, and the log source table notes they include session state and tool execution details. There is no accompanying warning about potentially sensitive contents, despite the skill encouraging direct viewing of these artifacts.

Ssd 3

Medium
Confidence
82% confidence
Finding
The skill includes a ready-made diagnostic pattern that reads session files and prints recent user message text, which is private user-provided content rather than operational metadata. Although framed as troubleshooting, the natural-language intent is to surface conversation contents from stored sessions, creating a data-leak risk that depends on semantic understanding of what those files contain.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.