The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses multiple command-line interfaces to interact with local and remote Drupal environments.
Evidence: Executes drush, git, ddev, terminus (Pantheon), acli (Acquia), platform (Platform.sh), upsun, and lagoon (amazee.io) commands across all files.
[DYNAMIC_EXECUTION]: The skill utilizes the drush php:eval command to perform programmatic updates to the Drupal database configuration, which involves executing PHP code snippets provided in the instructions.
Evidence: Found in SKILL.md and references/examples.md for updating config_split settings directly in the active configuration.
[INDIRECT_PROMPT_INJECTION]: The skill processes external Drupal configuration files (YAML), which creates a potential surface for instructions embedded in data to influence the agent's behavior during sync or merge operations.
Ingestion points: Reads configuration YAML files from the local filesystem (config/default/*.yml) and retrieves remote configuration via hosting CLIs as documented in SKILL.md and references/examples.md.
Boundary markers: The instructions do not specify delimiters or boundary markers to isolate the configuration data from the execution context.
Capability inventory: The skill has the capability to execute shell commands, perform file system writes, and interact with remote hosting platforms.
Sanitization: No specific sanitization or validation of the YAML content is mentioned before it is processed by the agent or passed to CLI tools.

drupal-config-mgmt