The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project source code and executes test suites, creating a surface where malicious code in the user's project could execute commands or influence agent behavior. * Ingestion points: Reads Rails models, controllers, and specification files during the refactoring process (e.g., spec/requests/orders_spec.rb mentioned in SKILL.md). * Boundary markers: The skill does not define specific delimiters or instructions to isolate the agent from potentially malicious code snippets in the files it modifies. * Capability inventory: The skill uses bundle exec rspec and bundle exec rubocop to execute code and analyze the project state. * Sanitization: No specific sanitization or validation of the ingested code is described.
[COMMAND_EXECUTION]: The skill instructions require the agent to execute shell commands to run test suites and linters as part of its mandatory verification protocol. * Evidence: Documented use of bundle exec rspec and bundle exec rubocop throughout SKILL.md and assets/complete_example.md to verify the state of the code after structural changes.

refactor-code