The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to execute the dreamina-canvas CLI tool to perform operations such as creating, editing, and upscaling image nodes.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input via the --prompt parameter, which is interpolated into shell commands. This creates a surface where malicious prompt content could attempt to influence the agent or the underlying system.
Ingestion points: User-provided text strings identified by the <prompt> placeholder in SKILL.md and references/image-node-contract.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands within the prompt are provided.
Capability inventory: Execution of the dreamina-canvas CLI tool for image generation and node management.
Sanitization: There are no instructions for sanitizing or escaping the user-provided prompt before it is passed to the command line.

dreamina-canvas-generate-image