Back to skill

Security audit

HTML Animation Video

Security checks for vulnerabilities and agentic risk

Overview

This skill creates animations as described, though optional video rendering can install software and overwrite an existing video.

Use a fresh project folder to protect existing output.mp4 files. For optional rendering, prefer pinned, project-local dependencies and control software installation separately. A manual-recording fallback is provided. All bundled files were reviewed, including the scanner's partial-coverage areas; external downloads were not inspected.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (5)

The rendering workflow directs automatic dependency installation without warning about environment changes.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The workflow directs the agent to install missing dependencies without distinguishing permission to render a video from permission to modify the execution environment. Although these are legitimate rendering tools, automatic installation can execute downloaded package code and change shared tooling without the user's informed approval.

Content

No source excerpt is available for this finding.

The particle-animation style requires an English conclusion subtitle without user opt-in.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The required visual elements include “一句英文结论字幕” (a concluding subtitle in English). This explicitly mandates a natural language for generated output without offering a language choice or documenting a justification for that restriction.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx playwright'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

The MP4 rendering command silently overwrites an existing output.mp4 without warning or confirmation.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The ffmpeg command combines the overwrite flag '-y' with the fixed filename 'output.mp4', so an existing file in the working directory can be replaced without confirmation. Rendering is a legitimate operation, but permission to create a new video does not imply permission to destroy a previous output.

Content

No source excerpt is available for this finding.

Standalone style and technical terms are listed as activation triggers without an animation-request constraint.

Low
Category
Trigger Abuse
Confidence
97% confidence
Finding

The trigger list includes broad terms such as “ASCII,” “粒子” (particles), and “水墨” (ink painting), which can occur in unrelated programming, science, or art requests. It does not explicitly require an animation-creation request or provide exclusions, allowing unintended invocation.

Content

No source excerpt is available for this finding.