peach-help

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed exclusively as an onboarding and guidance assistant. It explicitly instructs the agent to remain read-only and avoid generating code or modifying files.
  • [COMMAND_EXECUTION]: The skill includes Bash in its allowed tools. Analysis of the instructions shows it is intended for project structure exploration, such as checking for the existence of documentation directories, which is consistent with its purpose as a guide.
  • [DATA_EXFILTRATION]: No network operations or access to sensitive credentials or environment files were detected. The skill maintains local scope for information gathering.
  • [PROMPT_INJECTION]: There are no attempts to override system instructions, bypass safety filters, or extract system prompts. The instructions are focused on task classification and helpful responses.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 04:09 PM