The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard Ruby on Rails development commands, including rails generate, rails db:migrate, and bundle exec rspec, to automate routine development tasks and validation steps.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data such as user requirements and existing application source code, which represents a potential attack surface for indirect prompt injection.
Ingestion points: User-provided requirements, model definitions, and controller logic processed during the implementation workflow.
Boundary markers: The skill does not explicitly define delimiters to isolate untrusted user requirements from the agent's internal instructions.
Capability inventory: The agent has the capability to execute shell commands (rails, bundle) and modify project files based on its analysis.
Sanitization: The skill correctly mandates security best practices for database interactions, such as using sanitize_sql and parameterized queries, although it lacks specific prompt-level isolation for general data ingestion.

rails-expert