The Agent Skills Directory

[PROMPT_INJECTION]: The skill defines a workflow for creating code modules based on user requirements, which establishes a surface for indirect prompt injection.
Ingestion points: User-provided feature descriptions and architecture requirements processed through the instructions in SKILL.md.
Boundary markers: The skill lacks specific delimiters or instructions to ignore potential commands within user-provided data.
Capability inventory: The skill instructs the agent to perform file-system operations and execute shell commands (e.g., ripgrep, Gradle) as detailed in references/validation.md.
Sanitization: No explicit sanitization or input validation for content derived from user feature requests is described.
[COMMAND_EXECUTION]: The skill provides templates for using shell utilities such as rg (ripgrep) to check for architectural leaks and ./gradlew to run build tasks. These operations are intended for use within the project directory to support development workflows.
[EXTERNAL_DOWNLOADS]: The skill includes references to official documentation and technical guides from trusted sources like Google's Android developer site and the Gradle documentation project for educational and reference purposes.

modular-architecture