The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions specify running a local script named validate_story_project.py to verify the narrative state. This is a standard project-specific validation step and does not involve remote code execution or privilege escalation.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and process untrusted narrative data (chapters and screenplay scenes) which could contain embedded instructions intended to influence the agent's behavior during the state update process.
Ingestion points: Narrative content from NARRATIVE_STATE.md and external chapter/scene files specified in the workflow (SKILL.md).
Boundary markers: None identified; the skill does not explicitly instruct the agent to ignore instructions embedded within the narrative text.
Capability inventory: The skill has the capability to read local files, update the story_state.json file, and execute the validate_story_project.py script.
Sanitization: There are no explicit instructions for sanitizing or escaping the content retrieved from the narrative files before processing.

story-state