The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted user data (story ideas, scenarios, and files) through a multi-agent pipeline.
Ingestion points: User-provided topics, story ideas, and existing scenario/character files are extracted and saved to _workspace/00_input.md in Phase 1.
Boundary markers: The skill does not explicitly define delimiters or instructions for the agents to ignore potential instructions embedded within the user-provided comic scenarios.
Capability inventory: The agents have capabilities to write files to the local workspace (_workspace/), communicate via SendMessage, and trigger image generation via Gemini.
Sanitization: No explicit sanitization or validation of user-provided story content is documented before it is passed to the storyboarder or dialogue-writer agents.
[NO_CODE]: This skill consists entirely of high-level instructions and configuration for an agent team. It does not contain executable scripts, shell commands, or external dependencies, significantly reducing the technical attack surface.

comic-creator