new-rails-project

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is a generator that translates user requirements into shell commands, creating an attack surface for indirect prompt injection.
  • Ingestion points: The agent is instructed to ask numerous clarifying questions via the AskUserQuestionTool to gather project requirements (SKILL.md).
  • Boundary markers: The skill includes a 'Limitations' section advising against treating output as a substitute for expert review, but it lacks technical boundary markers or instructions to ignore instructions embedded in user input.
  • Capability inventory: The skill utilizes the Bash tool with wide permissions for rails, bundle, npm, and yarn commands (SKILL.md).
  • Sanitization: The instructions do not define any validation or sanitization for the project name ($1) or user-supplied requirements before interpolation into shell commands.
  • [SAFE]: The skill incorporates security-positive instructions such as running Brakeman for security scanning and using encrypted fields for sensitive data like OAuth tokens.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:07 AM
Security Audit — agent-trust-hub — new-rails-project