The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, including story text and scene JSON, to generate prompts for image generation. This creates a surface for indirect prompt injection where malicious instructions embedded in the input text could influence the agent's behavior. \n
Ingestion points: {slug}_scenes.json and user-provided story text (SKILL.md). \n
Boundary markers: None identified; instructions are interpolated directly into prompts. \n
Capability inventory: The skill executes a local Python script (fal_image.py) which performs network requests to the Fal API and writes image files to the local filesystem (SKILL.md, assets/fal_image.py). \n
Sanitization: No explicit sanitization or filtering of input data is described before prompt construction. \n- [COMMAND_EXECUTION]: The skill is designed to execute a bundled Python utility script, assets/fal_image.py, to handle image generation tasks. This is a legitimate and expected behavior for the skill's primary function. \n- [EXTERNAL_DOWNLOADS]: The fal_image.py script downloads generated images from Fal AI's CDN (v3.fal.media). This is a well-known service and the operation is necessary for the skill's narrative illustration purpose.

story-illustrator