The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses ddev exec, ddev drush, and ddev export-db to manage the Drupal environment. These commands are executed within a DDEV container or from the host project root to perform operations like database exports, applying recipes, and running Drupal CLI tools (vendor/bin/dr). This behavior is consistent with the skill's primary purpose of managing a development environment.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and review Drupal recipes and configuration files (YAML format). This creates a surface for indirect prompt injection where untrusted data could potentially influence the agent's reasoning.
Ingestion points: Drupal recipe files (recipe.yml), configuration files (config/*.yml), and content files (content/*.yml) are read from the local filesystem.
Boundary markers: The skill does not explicitly define markers or headers to isolate untrusted file content from the agent's core instructions.
Capability inventory: The agent is empowered to execute shell commands via ddev, modify the filesystem, and perform database operations.
Sanitization: Instructions rely on standard YAML parsing and manual review, with no specific automated sanitization logic mentioned for external content.

drupal-recipes