The Agent Skills Directory

[SAFE]: The skill instructions establish clear safety boundaries, explicitly forbidding the agent from running the project's Ruby code.
[SAFE]: The command-line operations mentioned (e.g., bundle exec srb tc, mise exec) are industry-standard tools for the Ruby on Rails development ecosystem used for type checking and environment management.
[SAFE]: The large collection of referenced Ruby files consists of legitimate parameter and resource definitions from the well-known stripe-ruby library, provided as context for the agent to understand complex type structures.
[SAFE]: No data exfiltration, credential harvesting, or unauthorized network access patterns were detected in the instructions or the referenced code files.
[SAFE]: The skill uses a structured approach (checklists and tracking files) which maintains a consistent state and reduces the risk of accidental misconfiguration during automated processing.

generating-sorbet