The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill interacts with local repository files, creating a potential surface for indirect injection if project files contain malicious instructions.
Ingestion points: The skill reads db/schema.rb and existing migration files within the db/migrate/ directory to determine the current database state.
Boundary markers: None present; the skill treats the content of these files as standard technical data.
Capability inventory: The skill has the ability to write new files and execute shell commands via bin/rails tasks.
Sanitization: No specific sanitization or filtering of the repository's file content is performed before processing.
[DYNAMIC_EXECUTION]: The skill is designed to generate Ruby migration scripts and execute them through the bin/rails db:migrate interface. While this involves generating and running code, it is the primary intended function of a migration assistant and follows standard development workflows.
[COMMAND_EXECUTION]: The skill requests access to execute a specific subset of shell commands related to the Ruby on Rails framework, including migration generation and database management tasks. These operations are essential for the skill's utility in a software development environment.

migrations