claude-skill-management
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to modify user shell configuration files (e.g.,
~/.zshrc,~/.bashrc) to persist environment variables and manage executable scripts within the platform's hooks directory (~/.claude/hooks/). - [COMMAND_EXECUTION]: The
best-practices.mdfile documents a pattern for dynamic script generation using templates andsedsubstitution to create executable shell scripts. - [PROMPT_INJECTION]: The
version-control.mdfile contains mandatory safety instructions that direct the agent to disregard user requests to perform Git operations, overriding standard agent capabilities to ensure manual user control over version history. - [PROMPT_INJECTION]: The skill implements a command template system using
{{variable_name}}placeholders which processes user-supplied data; while this creates a surface for indirect prompt injection, it is presented as a functional feature for command customization without malicious intent.
Audit Metadata