generators
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external project data which presents a surface for indirect prompt injection.
- Ingestion points: The skill reads config/application.rb and generator templates in lib/templates/ (referenced in SKILL.md and references/templates.md) to match project conventions.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are present when reading project files.
- Capability inventory: The skill can write/edit files and execute scoped Rails commands as defined in the SKILL.md frontmatter (bin/rails generate, bin/rails destroy).
- Sanitization: No explicit sanitization or validation logic is defined for the content ingested from the project files.
- [REMOTE_CODE_EXECUTION]: The skill documents standard Rails functionality for downloading and applying remote application templates.
- Evidence: The documentation in references/templates.md describes how to use rails new -m and bin/rails app:template with external URLs.
- Context: The examples use the reserved example.com domain, and the functionality is a core feature of the Rails framework used for automation.
- [DYNAMIC_EXECUTION]: The skill documents Thor actions used for file manipulation and running commands during generation.
- Evidence: references/thor-actions.md lists methods like run, rails_command, and rake for executing shell commands.
- Context: These are standard internal methods for Rails generators and are presented as part of the API documentation for creating custom generators.
Audit Metadata