The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from external files within the .ddd/obligations/ directory to generate automated gates. This creates a surface where instructions embedded in project documentation could influence agent behavior. Ingestion points: Obligation definitions are read from the .ddd/obligations/ directory as described in Step 1. Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the processed data. Capability inventory: The skill directs the agent to generate executable configuration files (e.g., for dependency-cruiser or eslint) and execute validation fixtures (Step 3 and Step 5). Sanitization: There is no mention of input validation or escaping for the data used to generate these controls.
[DYNAMIC_EXECUTION]: The skill involves the generation of executable configurations and test scripts (e.g., .dependency-cruiser.js) based on documentation (Step 1 and Step 4a). The creation of executable logic that is integrated into a project's CI/CD pipeline represents a dynamic execution risk.

ddd-controls