Trust Assessment
sag received a trust score of 83/100, 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, 1 high, 1 medium, and 0 low severity. Key findings include Potential Command Injection via CLI Tool, Unpinned External Dependency (Supply Chain Risk).
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 18, 2026 (commit b62bd290). 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 via CLI Tool The skill documentation demonstrates the use of the `sag` command-line interface (CLI) tool, which will be executed by the agent. If the agent constructs these shell commands by directly interpolating untrusted user input without proper sanitization (e.g., quoting or escaping), it could lead to command injection. An attacker could craft malicious input that executes arbitrary commands on the host system. The agent implementation must ensure that all user-provided input passed to the `sag` command (or any other shell command) is rigorously sanitized and properly quoted/escaped to prevent arbitrary command execution. For example, using `shlex.quote()` in Python or similar functions in other languages is recommended when constructing shell commands. | LLM | SKILL.md:29 | |
| MEDIUM | Unpinned External Dependency (Supply Chain Risk) The skill's manifest specifies installation via a Homebrew formula (`steipete/tap/sag`) without pinning a specific version. This means that any update to the `steipete/tap/sag` formula could introduce breaking changes, vulnerabilities, or even malicious code without explicit review. This constitutes a supply chain risk, as the integrity of the skill depends on the upstream maintainer of the Homebrew tap. If possible, specify a pinned version for the `sag` binary in the installation instructions (e.g., `sag@1.2.3`). Regularly review the upstream Homebrew formula for changes. Consider vendoring critical dependencies or using a more controlled installation method if the risk is deemed too high. | LLM | manifest.json:1 |
Scan History
Embed Code
[](https://skillshield.io/report/f0256833a556832c)
Powered by SkillShield