ClawHub

WayinVideo - AI Clipping

Security checks across malware telemetry and agentic risk

Overview

This is a coherent WayinVideo clipping integration, but users should understand that videos or URLs may be sent to WayinVideo and result files are saved locally.

Install only if you are comfortable using a WayinVideo API key and sending the selected video file or URL to WayinVideo for processing. Keep generated JSON result files and temporary export links private, and avoid enabling progress system events if you do not want background notifications.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
Findings (6)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
def send_system_event(text):
    try:
        subprocess.run([
            "openclaw", "system", "event",
            "--text", text,
            "--mode", "now"
Confidence
93% confidence
Finding
subprocess.run([ "openclaw", "system", "event", "--text", text, "--mode", "now" ], check=True, capture_output=True, text=True)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill invokes Python scripts, reads environment variables, accepts local file paths, writes result files, and sends data to an external API, but it does not declare corresponding permissions. This creates a transparency and policy-enforcement gap: users and hosting frameworks may not realize the skill can access local files, exfiltrate video content/URLs, and run shell commands.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The script has a side capability to emit host-level system events, which is not necessary for polling clip results. That expands the skill's authority beyond its declared purpose and could be used to spam, socially engineer, or otherwise manipulate the host/user with notifications triggered by remote service state.

Vague Triggers

Medium
Confidence
79% confidence
Finding
The description contains broad invocation language such as highlight extraction and best moments, which can cause the skill to trigger in contexts where the user did not specifically intend to use an external video-processing service. In this skill, overbroad routing is more dangerous because invocation may lead to local file handling, API-key use, uploads to a third party, and background polling.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs the agent to upload local files or process URLs through the WayinVideo API but does not present a clear user-facing warning that video content, metadata, and possibly sensitive local media will be transmitted to a third-party service. This is a real privacy and data-governance issue, especially because unsupported URLs are to be downloaded first and then uploaded, increasing the chance of unintentionally sending private or copyrighted material off-platform.

Unvalidated Output Injection

High
Category
Output Handling
Content
def send_system_event(text):
    try:
        subprocess.run([
            "openclaw", "system", "event",
            "--text", text,
            "--mode", "now"
Confidence
90% confidence
Finding
subprocess.run([ "openclaw", "system", "event", "--text", text, "--mode", "now" ], check=True, capture_output

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal