claude-code-plugin-reference
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists exclusively of technical documentation intended to provide the AI agent with architectural context. It contains no malicious patterns, prompt injections, or exfiltration vectors.
- [NO_CODE]: The skill includes only Markdown files (
SKILL.md,modules/*.md) and does not distribute any scripts, binaries, or configuration files that execute code. - [DYNAMIC_EXECUTION]: Although the documentation explains how the platform handles dynamic elements like hooks and JavaScript workflows, the skill itself does not implement or trigger any dynamic execution.
- [DATA_EXPOSURE_AND_EXFILTRATION]: Evaluated for credential harvesting and data access; the documentation correctly instructs developers to use environment variables and secure paths, but does not access any sensitive local data itself.
- [COMMAND_EXECUTION]: Technical examples (e.g.,
uv run python) are provided for instructional purposes only and are not formatted as autonomous agent commands.
Audit Metadata