The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill generates templates that reference official base images from well-known registries including Docker Hub and Microsoft Container Registry (e.g., node:22-bookworm-slim, python:3.12-slim-bookworm, postgres:16-alpine, mcr.microsoft.com/dotnet/sdk). These references are consistent with the skill's primary purpose of application containerization.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of user-provided application requirements to generate configuration files. This presents a potential surface for indirect prompt injection if the user provides malicious strings as input.
Ingestion points: User input collected during Step 1: Requirements Gathering in SKILL.md.
Boundary markers: Absent. The instructions do not specify the use of delimiters or escaping mechanisms when interpolating user data into templates.
Capability inventory: The agent is instructed to write files to the filesystem (Dockerfile, compose.yaml, .dockerignore, .env.example).
Sanitization: Absent. No specific validation or sanitization of user-provided strings (such as app names, ports, or environment variables) is performed before they are placed into the generated files.
[CREDENTIALS_UNSAFE]: The skill includes security-conscious examples for handling secrets, such as generating .env.example files with placeholder values (e.g., POSTGRES_PASSWORD=changeme) and warning against committing real secrets. It explicitly advises against using secrets in ENV or ARG layers.

docker-agents-generator