The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from user-controlled files including series.md, characters.md, and episode-card.md to generate video prompts. While this represents a potential attack surface for indirect prompt injection, the skill's capabilities are limited to markdown text generation.
Ingestion points: series.md, characters.md, episodes/epXXX/episode-card.md (specified in SKILL.md).
Boundary markers: The instructions do not define specific delimiters or warnings to separate user data from agent instructions.
Capability inventory: The skill is restricted to file reading and generating markdown output; it lacks network access, file-writing capabilities (beyond its own output), or command execution as per the frontmatter and stop conditions.
Sanitization: No explicit sanitization or filtering of external content is described in the workflow.
[COMMAND_EXECUTION]: The skill templates in templates/video-prompts-en.md reference CLI-style commands such as multimodal2video and --prompt flags. However, these are strictly documentation components intended for a downstream adapter (plotloom-video-adapter) and are not executed by the skill itself. The skill instructions explicitly mandate stopping before calling any models or submitting jobs.

plotloom-shot-prompts