shell-scripting
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts on user-defined shell tasks and existing script content, creating an attack surface where untrusted input could attempt to influence the logic of generated shell code.
- Ingestion points: User-provided tasks or shell snippets (defined as the
<task-or-shell>argument in SKILL.md). - Boundary markers: The skill utilizes markdown code blocks for generated shell scripts, which provides clear structural boundaries.
- Capability inventory: The skill is capable of generating shell commands and modifying shell configuration files (e.g., .bashrc, .zshrc) as detailed in
references/bash.md,references/zsh.md, andreferences/ssh-tmux-autostart.md. - Sanitization: The skill relies on the agent's internal safety filters and the specific instruction to 'detect the target before writing code' rather than implementing custom sanitization for user inputs.
Audit Metadata