Security audit

视频

Security checks for vulnerabilities and agentic risk

Overview

This video-generation skill should be reviewed because it can direct an agent to run local commands, including a custom TTS command, without clear safeguards.

Install only if you intend to let the agent run local media-generation commands. Avoid using --tts-command with untrusted script text or command strings, verify what generate.py will be executed from your environment, and keep output paths limited to a working directory you control.

A.I.G

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

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (3)

Intent-Code Divergence

High

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill advertises command execution as occurring in a 'safe sandbox', yet elsewhere explicitly supports a user-controlled --tts-command template that can invoke arbitrary shell commands. This creates a dangerous mismatch: users or downstream agents may trust the safety claim and pass untrusted content into a feature that can execute attacker-chosen commands, leading to command injection or arbitrary code execution in the agent environment.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill description uses very broad activation language around video processing, audio editing, media conversion, and voice generation, which can cause an agent to invoke this skill for many loosely related requests. In a skill that includes exec and file-writing capabilities, overbroad triggering increases the chance that the tool is used in unintended contexts with sensitive local files or untrusted inputs.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The documentation provides direct examples of running shell commands and writing output files, including a customizable --tts-command, without prominent warnings that these operations can execute local programs and modify the filesystem. In an agent setting, this omission increases the risk that users or automated systems treat these actions as routine and pass attacker-influenced values into command or path parameters.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.