Claude Code Guide
Pass
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: LOW
Full Analysis
- [NO_CODE] (SAFE): The skill consists entirely of Markdown documentation and configuration templates. It contains no executable scripts, binary files, or automation that would execute on a host system.
- [PROMPT_INJECTION] (LOW/INFO): The guide mentions phrases such as "forget previous instructions." While these are common injection triggers, they are used here in an instructional context for manual session management and debugging by the user, rather than an attempt to bypass the agent's safety protocols.
- [EXTERNAL_DOWNLOADS] (LOW/INFO): The documentation links to an external GitHub repository (
github.com/zebbern/claude-code-guide) for reference. No automated downloading, cloning, or execution of remote code is performed by the skill.
Audit Metadata