ClawHub
Back to skill

Security audit

workflow-monitor

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent workflow-monitoring helper, but it can capture command output and session context and post them into GitHub/GitLab issues without clear redaction safeguards.

Install only if you are comfortable with workflow diagnostics being captured and potentially filed into your issue tracker. Keep auto_create_issues disabled unless the repository is trusted and private, and review/redact command output, paths, session IDs, environment notes, tokens, stack traces, and proprietary snippets before allowing issue creation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes very generic terms like "workflow," "monitoring," and "automation," which can cause the skill to activate during ordinary discussion rather than explicit user intent. In a skill that can analyze sessions and file issues on a git platform, accidental invocation increases the chance of unintended data collection or unwanted issue creation.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly describes logging execution events, storing evidence, and linking command outputs and session information into issues, but it does not provide a clear user-facing warning about sensitive data exposure. Workflow logs and command output frequently contain secrets, tokens, internal paths, stack traces, or proprietary code fragments, so automatically persisting and reposting this material can leak confidential information to issue trackers or other collaborators.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The rate-limiting section permits automatic issue creation via a broad `auto_create_issues: true` toggle without defining stricter eligibility criteria, approval scope, or safeguards on what content may be filed. In a workflow-monitoring skill that can ingest noisy or attacker-influenced command/output data, this can lead to unwanted issue spam and automated publication of sensitive diagnostic details.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The error-report template explicitly embeds raw command text, output excerpts, working directory, environment notes, workflow name, and session identifier into GitHub issues with no warning, minimization, or sanitization step. Since workflow failures often include secrets, tokens, internal paths, hostnames, or proprietary data, this creates a direct risk of sensitive information disclosure to anyone who can view the issue tracker.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.