The Agent Skills Directory

[SAFE]: The skill provides legitimate architectural guidance for Ruby on Rails 7 applications, covering Active Record, Hotwire, Sidekiq, and RSpec.
[PROMPT_INJECTION]: No evidence of instructions designed to override system prompts or bypass safety guardrails. The instructions are focused on software engineering tasks.
[DATA_EXFILTRATION]: No unauthorized data access or network exfiltration patterns. It correctly references the Rails credential system (Rails.application.credentials.secret_key_base) rather than hardcoding secrets.
[COMMAND_EXECUTION]: The workflow utilizes standard development commands (rails, bundle, rspec). These are expected in the context of a code-generation and development-assistance skill.
[EXTERNAL_DOWNLOADS]: References well-known, reputable Ruby gems (e.g., sidekiq, jwt, rack-attack) from the official registry. No use of untrusted scripts or remote execution from unknown sources.
[CREDENTIALS_UNSAFE]: The skill instructs the user to use parameterized queries and strong_parameters to prevent SQL injection and mass assignment vulnerabilities, demonstrating a security-first approach.

rails-expert