extension-authoring-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructional content and templates for extension development. Analysis of all six files confirms the content is educational and functional without any hidden malicious instructions.
- [SAFE]: No prompt injection or safety bypass patterns were identified. The instructions guide the user to follow structured XML and YAML patterns consistent with the platform's standard operating procedures.
- [SAFE]: The skill emphasizes security best practices in its reference materials, such as implementing tool restrictions via
allowed-tools, preventing infinite loops in hooks, and maintaining path safety for scripts. - [SAFE]: Data operations are limited to local metrics logging (e.g., writing to
~/.claude/skill-analytics/) and standard tool usage (e.g.,git status,jq), which are appropriate for a developer productivity tool. - [SAFE]: No obfuscation, multi-layer encoding, or suspicious remote dependencies were found.
Audit Metadata