The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown documentation and code templates. No executable code is distributed with the skill, ensuring it cannot perform autonomous actions or unauthorized operations.
[EXTERNAL_DOWNLOADS]: The templates reference standard Ruby gems such as rspec, factory_bot, and webmock. These are widely used, legitimate packages from the official RubyGems registry, documented here for project setup purposes.
[COMMAND_EXECUTION]: Instructions provide standard CLI commands for Ruby development (e.g., bundle exec rspec, gem install). These are routine developer actions and do not involve suspicious or obfuscated command strings.
[INDIRECT_PROMPT_INJECTION]: The skill provides patterns for generating test code based on user-provided application logic. This surface is well-contained as the skill does not execute the generated code itself.
Ingestion points: User descriptions of application features or Ruby code in the conversation context.
Boundary markers: Examples demonstrate the use of standard Markdown blocks to delineate code.
Capability inventory: No internal execution capabilities; functionality is limited to providing text-based templates.
Sanitization: Relies on the host AI platform's inherent safety guardrails for code generation.

rspec-skill