The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard Rails development commands including bundle exec rspec for testing, bundle exec rubocop for linting, and bin/rails console for verification. These commands are typical for the described workflow and are restricted to the local environment.
[SAFE]: The skill and its associated documentation follow industry-standard patterns for Ruby on Rails development. There are no detections of credential theft, obfuscation, or unauthorized external communications.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external project files to generate code. Ingestion points: Project source code including models, queries, and jobs. Boundary markers: None defined in the skill instructions. Capability inventory: Ability to create and modify Ruby files in the app/services/ directory and execute shell commands via bundle exec. Sanitization: Code examples demonstrate defensive practices such as the use of strong parameters and explicit authorization guards.

service-agent