The Agent Skills Directory

[SAFE]: The skill consists of legitimate technical documentation and templates for Coolify deployments, adhering to industry-standard practices for containerized applications.
[INDIRECT_PROMPT_INJECTION]: The skill guides the agent in generating deployment artifacts based on user-provided application metadata and environment variables.
Ingestion points: User-supplied environment variables (e.g., DATABASE_URL) and repository file structures described in SKILL.md.
Boundary markers: None explicitly defined in the provided templates.
Capability inventory: File system writes (Dockerfiles, entrypoint scripts) and terminal execution of setup commands (database creation scripts) within the container context.
Sanitization: The skill provides logic to normalize connection strings (e.g., replacing postgres:// with postgresql://) but otherwise relies on the user to provide valid inputs. The risk is considered low and inherent to the intended deployment automation functionality.
[COMMAND_EXECUTION]: Provides standard deployment commands and health check snippets using tools like curl, wget, npm, and pip. These are executed within the user's defined build or runtime environment.
[DYNAMIC_EXECUTION]: Includes inline Node.js and Python scripts for database bootstrapping and health validation during the container startup process, which is a common and accepted pattern for PaaS deployments.

coolify-deploy