ClawHub
Back to skill

Security audit

动态漫画制作

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent motion-comic production skill whose file, command, image-generation, and online TTS behavior matches its stated purpose, with privacy and input-validation caveats.

Install only if you are comfortable with Codex creating project files, running FFmpeg/ffprobe and helper scripts, and sending dialogue text to Edge TTS online. Use a dedicated project directory, review TTS manifests from untrusted sources before running them, and avoid using sensitive scripts with the default online TTS route.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
)
    concat_file = out / "voice-concat.txt"
    concat_file.write_text(concat_text, encoding="utf-8")
    subprocess.run(
        [
            "ffmpeg", "-hide_banner", "-loglevel", "error", "-y",
            "-f", "concat", "-safe", "0", "-i", str(concat_file),
Confidence
87% confidence
Finding
The code generates an ffmpeg concat manifest from user-influenced paths and then runs ffmpeg with '-safe 0', which disables path safety checks and permits absolute and otherwise unsafe paths. If an attacker can control manifest line IDs or the audio directory contents, they may cause ffmpeg to read unintended local files or special file paths during media processing, which is more dangerous in an automation skill that processes external assets.

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs the agent to read and write project files and execute shell commands such as FFmpeg, ffprobe, and local scripts, but it does not declare corresponding permissions or constraints. This creates a capability-transparency gap: users and policy layers may not realize the skill can modify files or run commands, increasing the chance of unintended filesystem changes or command execution in a powerful local environment.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly relies on Edge TTS via an online Microsoft speech endpoint and notes that it requires network access, but it does not clearly warn users that script content and dialogue will be transmitted to a third-party service. In a motion-comic workflow, scripts may contain unpublished creative work, personal data, or sensitive business content, so silent transmission creates a real privacy and data-handling risk.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The script sends `line['text']` and voice parameters to `edge-tts`, which typically relies on a remote Microsoft Edge TTS service. If manifests can contain private scripts, unpublished content, or sensitive user data, this creates an information disclosure risk because data is transmitted to an external service without explicit consent, warning, or documented controls.

Static analysis

No suspicious patterns detected.