ClawHub
Back to skill

Security audit

openclaw-audit-watchdog

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed audit watchdog that schedules OpenClaw security reports to configured recipients, with notable but purpose-aligned persistence and delivery behavior.

Install this only on hosts where you want unattended recurring OpenClaw security audit reports. Before enabling it, verify the DM channel, recipient, optional email address, SMTP/sendmail settings, install directory, and schedule; remove or disable the cron job if you only wanted a one-time audit.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
87% confidence
Finding
The skill documentation claims it will create/update a recurring cron audit workflow with DM/email delivery, but the static finding indicates the actual behavior diverges materially, including unrelated code review functionality, optional repository updates, and missing persistence or DM implementation. In a security automation skill, this mismatch is dangerous because operators may grant trust, schedule execution, or deploy it fleet-wide under false assumptions about what it does and does not do.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes very generic phrases such as "security audit," "run audit," and "security report," which are likely to overlap with normal user requests and can cause unintended invocation of a high-impact skill. In this skill’s context, accidental activation is more dangerous because the skill can perform recurring setup, send reports externally, and establish persistence via cron, so a mistaken trigger may lead to unwanted execution and data disclosure.

Session Persistence

Medium
Category
Rogue Agent
Content
## Goal

Create (or update) a daily cron job that:

1) Runs:
- `openclaw security audit --json`
Confidence
89% confidence
Finding
Create (or update) a daily cron job that: 1) Runs: - `openclaw security audit --json` - `openclaw security audit --deep --json` 2) Summarizes findings (critical/warn/info + top findings) 3) Sends t

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.