Trust Assessment
create-adr 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 Skill requests broad file system read/write access.
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 June 1, 2026 (commit 81e7e0dd). 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 | Skill requests broad file system read/write access The skill explicitly describes a workflow that requires reading directory contents to determine the next sequential ADR number and writing files to a user-specified location. This implies broad file system read (directory listing) and write capabilities. Without strict sandboxing, this could lead to data exfiltration (e.g., listing sensitive directories, or if prompted, reading file contents) or unauthorized file modification/creation outside the intended scope. The skill also allows the user to specify 'a different location' for saving, which further broadens the potential write scope. Implement strict sandboxing for file system operations. Limit directory scanning to specific, pre-approved paths. Restrict file writing to a designated, isolated output directory. Avoid allowing the LLM to determine arbitrary file paths for writing. Consider using a dedicated tool or API for file system interaction that enforces these limits, rather than relying on the LLM's direct interpretation. | LLM | SKILL.md:160 |
Scan History
Embed Code
[](https://skillshield.io/report/fd5d8fe96b1132ad)
Powered by SkillShield