shell-review

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements standard security and portability auditing patterns for shell scripts. It leverages common system utilities (find, grep, rg) and recommends the use of industry-standard tools like ShellCheck. All operations are confined to the local workspace and align with the skill's stated purpose.
  • [PROMPT_INJECTION]: The skill has an inherent indirect prompt injection surface because its primary function is to read and analyze untrusted shell scripts from the local environment.
  • Ingestion points: The skill reads shell scripts (files with .sh extension or shebangs) found in the user's workspace for analysis.
  • Boundary markers: There are no explicit instructions in the skill to treat input code as untrusted or to ignore instructions embedded within comments in the audited scripts.
  • Capability inventory: The skill utilizes the Read, Grep, and Bash tools to identify and examine script content.
  • Sanitization: No automated sanitization or escaping of script content is performed before the agent processes it for review.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 07:31 AM