Trust Assessment
dex received a trust score of 76/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 2 findings: 0 critical, 2 high, 0 medium, and 0 low severity. Key findings include Potential Command Injection in 'dex' tool arguments, Potential Data Exfiltration through 'dex' file operations.
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 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Potential Command Injection in 'dex' tool arguments The skill describes a 'dex' command-line tool that accepts various user-provided arguments (e.g., `-d`, `--context`, `--result`, `<id>`). If the underlying implementation of the 'dex' tool does not properly sanitize or escape these arguments before executing them in a shell context, an attacker could inject arbitrary shell commands. This could lead to remote code execution or other malicious activities. Implement robust input sanitization and escaping for all user-provided arguments passed to the 'dex' tool. Avoid direct interpolation of user input into shell commands. Use parameterized commands or libraries that handle escaping automatically. If 'dex' is implemented in Python, use `subprocess.run` with `shell=False` and pass arguments as a list. | LLM | SKILL.md:10 | |
| HIGH | Potential Data Exfiltration through 'dex' file operations The skill explicitly states that tasks are stored as JSON files in `.dex/tasks/{id}.json` and provides commands like `dex show <id>` to view task content. If the `<id>` parameter is vulnerable to path traversal (e.g., `../../../../etc/passwd`) or if command injection (as described above) allows arbitrary file reads, an attacker could exfiltrate sensitive files from the system. Ensure that all file paths constructed from user input (like `<id>`) are strictly validated and sanitized to prevent path traversal. Restrict file operations to the intended `.dex/tasks/` directory using chroot or similar mechanisms if possible. Implement robust input validation for all arguments. | LLM | SKILL.md:13 |
Scan History
Embed Code
[](https://skillshield.io/report/ab3cd2ae21474d01)
Powered by SkillShield