ClawHub
Back to skill

Security audit

Wjs Editing Multicam

Security checks across malware telemetry and agentic risk

Overview

This skill is a local multicamera video-editing helper; its file access and ffmpeg execution fit that purpose, but users should choose output paths carefully because renders can overwrite files.

Install only if you intend to run local video-editing scripts on selected media files. Confirm the input videos, sidecars, and especially the EDL/MP4 output filenames before running, because existing files at those output paths may be overwritten; also make sure ffmpeg and numpy are available.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill documentation clearly describes reading source media and sidecar files, writing EDL/output files, and invoking ffmpeg/render scripts, which implies file_read, file_write, and shell capabilities. If the platform relies on declared permissions for sandboxing or user consent, omitting them creates a capability mismatch that can lead to over-broad execution or insufficient review of a shell-capable skill.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The trigger phrases include broad Chinese requests like '做个剪辑', '切几个机位', and '把这几个视频合成一个', which can match ordinary editing requests beyond this skill's narrow prerequisites. Because this skill performs file operations and shell-based rendering, overly broad activation increases the chance it is invoked in the wrong context, causing unintended access to local media files or execution of editing pipelines the user did not specifically request.

Missing User Warnings

Low
Confidence
84% confidence
Finding
The script writes JSON directly to a user-specified output path with `Path.write_text()` and does not check whether the destination already exists or require an explicit overwrite flag. In an agentic context, this can unintentionally clobber existing files if the caller supplies a sensitive or mistaken path, causing data loss or corruption.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The ffmpeg command includes -y, which forces overwrite of the output path without confirmation. In an agent context where output paths may be derived from automation or user-provided values, this can silently destroy existing files or clobber important media if the path is miscomputed or manipulated.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.