The Agent Skills Directory

[SAFE]: The skill provides a comprehensive guide for transforming standard Docker Compose files into Coolify-compatible templates. It explicitly instructs the agent to replace hardcoded passwords and sensitive credentials with Coolify's 'magic variables' (e.g., $SERVICE_PASSWORD_POSTGRES), which is a security-positive practice for secret management.
[SAFE]: No evidence of prompt injection, obfuscation, or malicious persistence mechanisms was found in the instructions, examples, or reference documentation.
[SAFE]: External links and references target well-known, legitimate services and open-source projects, including the official Coolify documentation and repositories for Ghost, Uptime Kuma, and Plausible Analytics.
[INDIRECT_PROMPT_INJECTION]: The skill involves processing user-provided Docker Compose files. While this creates a theoretical surface for indirect injection if a user provides a malicious compose file, the risk is minimal as the agent is instructed to perform specific, structural text transformations rather than executing the content.
Ingestion points: User-supplied YAML content in the chat context.
Boundary markers: None explicitly defined in the prompt instructions.
Capability inventory: Purely text transformation and generation; no system commands or network requests are triggered by the conversion process.
Sanitization: The conversion logic focuses on YAML structure and variable replacement, naturally ignoring or transforming most malicious payloads.

coolify-compose