The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill utilizes user-supplied text to populate prompts for image generation and editing tools, creating an indirect prompt injection surface.
Ingestion points: User prompts are ingested into the recommend and model_run tools.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the prompt logic.
Capability inventory: The skill possesses capabilities for network-based model execution (model_run), file writing/downloads (asset_download), and asset uploads (upload_asset).
Sanitization: No sanitization or validation of the user-provided prompt content is described.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to recommend an npx command (npx skills add scenario-labs/skills) to the user for installing related sibling skills. As this command is directed at the user and targets resources from the same author (scenario-labs), it represents standard vendor-provided functionality.

scenario-image