Trust Assessment
nb received a trust score of 86/100, 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 'nb git' subcommand.
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 Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Potential Command Injection via 'nb git' subcommand The `SKILL.md` explicitly states that the `nb git` subcommand can 'Run any git command'. This indicates that the `nb` tool acts as a passthrough for arbitrary `git` commands. If an LLM constructs `nb git` commands using untrusted user input without proper sanitization, an attacker could inject arbitrary shell commands. For example, `nb git 'status; rm -rf /'`. The `git` command itself is powerful and can interact with the filesystem, network, and execute hooks, leading to potential data loss, unauthorized access, or system compromise. 1. **LLM Integration Layer**: Implement robust input sanitization and validation for any user-provided arguments passed to `nb git` commands. Avoid directly concatenating untrusted user input into shell commands. Consider using an allowlist of safe `git` subcommands and arguments if possible. 2. **Skill Design**: If the full `nb git` passthrough functionality is not strictly necessary for the skill's intended use, consider restricting the `nb` tool's capabilities when used in an automated agent context, or wrap it with a safer interface that only exposes specific, safe `git` operations. 3. **Principle of Least Privilege**: Only expose the minimum necessary `git` commands required for the skill's functionality. | LLM | SKILL.md:168 |
Scan History
Embed Code
[](https://skillshield.io/report/32f14c383211588f)
Powered by SkillShield