ClawHub
Back to skill

Security audit

Video Creator

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed video-making helper that runs local media commands, with no hidden or deceptive behavior found.

Install this only if you want an agent to generate and run local video-processing commands. Review scripts before execution, avoid running examples in folders with important files, and only download or remix media you have rights to use.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

High
Confidence
94% confidence
Finding
The activation rule is extremely broad ('Use ALWAYS' plus many generic video-related phrases), which can cause this skill to trigger for loosely related requests and override more appropriate or safer skills. Over-broad routing increases the chance the agent will enter a code-execution workflow unnecessarily, expanding attack surface and causing unintended file handling or command generation in contexts that did not clearly request it.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The response format explicitly instructs the agent to 'Execute and deliver: Run the script and show the result' after generating bash or Python code, but provides no safety guardrails around local file creation, overwriting assets, networked tools, or user confirmation. In a skill centered on ffmpeg, Python, shell scripts, and optional download/transcription tools, this can directly lead to execution of unreviewed commands that modify the filesystem or process untrusted media inputs.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documented cleanup step uses a wildcard delete (`rm _seg_*.ts _concat_list.txt`) without any caution, path scoping, or safer deletion pattern. If a user runs the recipe from an unexpected directory, or if files matching that pattern already exist for another purpose, the command can cause unintended data loss.

Static analysis

No suspicious patterns detected.