The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local script scripts/screenplay_consistency.py to verify the generated Fountain screenplay against project metadata. This is a standard project-specific validation step.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from multiple project files to generate dialogue and narrative beats.
Ingestion points: Reads CHARACTER_PROFILE.md, 01_story/characters.md, 01_story/world.md, scene outlines, and adaptation maps.
Boundary markers: None explicitly defined to distinguish between data and instructions.
Capability inventory: Performs file writes to the filesystem and executes a Python verification script.
Sanitization: No validation or sanitization of the input story data is mentioned.

write-screenplay