Trust Assessment
local-vosk received a trust score of 88/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Potential Command Injection via User-Controlled Filename.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 14, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Potential Command Injection via User-Controlled Filename The skill's quick start guide demonstrates executing a script (`./skills/local-vosk/scripts/transcribe`) with user-controlled filenames (e.g., `voice_message.ogg`, `audio.mp3`) as arguments. If the `transcribe` script does not properly sanitize or escape these filenames before using them in internal shell commands (e.g., when calling `ffmpeg` or `vosk`'s CLI), an attacker could inject arbitrary shell commands by crafting a malicious filename. For example, a filename like `'; rm -rf /; #.ogg'` could lead to arbitrary code execution on the system where the skill is run. The `transcribe` script must rigorously sanitize and escape all user-provided input, especially filenames, before incorporating them into any shell commands. It is highly recommended to use libraries or functions designed for safe command execution (e.g., `subprocess.run` with `shell=False` and passing arguments as a list) to prevent shell injection vulnerabilities. | LLM | SKILL.md:18 |
Scan History
Embed Code
[](https://skillshield.io/report/cc05bac673df722a)
Powered by SkillShield