Security Audit
mattpocock/skills:skills/engineering/improve-codebase-architecture
github.com/mattpocock/skillsTrust Assessment
mattpocock/skills:skills/engineering/improve-codebase-architecture received a trust score of 85/100, 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 Command Injection via OS Command Execution on Generated File Paths.
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 July 17, 2026 (commit 9603c1cc). 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 | Command Injection via OS Command Execution on Generated File Paths The skill instructions direct the LLM to resolve the system temporary directory and execute OS-specific commands (`xdg-open <path>`, `open <path>`, `start <path>`) to open a dynamically generated HTML file. If the path resolution or filename generation incorporates untrusted input or is manipulated, this pattern can lead to arbitrary shell command execution. Avoid executing raw shell commands to open files. Instead, output the absolute path to the user and let them open it manually, or use safe, platform-agnostic APIs provided by the host environment that do not spawn a shell. | LLM | SKILL.md:41 |
Scan History
Embed Code
[](https://skillshield.io/report/dac30f096f1d87ae)
Powered by SkillShield