The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from local project files, creating a surface for indirect prompt injection.
Ingestion points: The scripts/detect-stack.sh script reads data from composer.json, package.json, .env, and other configuration files.
Boundary markers: The skill mitigates risks by presenting detection results for confirmation and showing previews before creating or modifying files.
Capability inventory: The skill can execute local scripts, write Docker configuration files, and run commands within containers using docker compose exec.
Sanitization: The skill relies on simple extraction via grep and requires manual confirmation from the user to validate all detected parameters.
[EXTERNAL_DOWNLOADS]: The skill downloads utility tools from trusted external sources.
Fetches wp-cli.phar from the official GitHub repository for WordPress (raw.githubusercontent.com) as part of the WordPress setup workflow.
Fetches the Drush launcher (drush.phar) from its official GitHub repository for Drupal environment configuration.
[COMMAND_EXECUTION]: The skill generates and runs configuration logic as a core part of its functionality.
Dynamically constructs Dockerfile and docker-compose.yml files at runtime based on embedded templates and user-specific input.
Executes verification and health check scripts (scripts/health-check.sh, scripts/db-test.sh) to ensure generated services are running as expected.

docker-local-dev