The Agent Skills Directory

[COMMAND_EXECUTION]: The workflow involves running a project-specific Python script, scripts/character_profiles.py, and a project validator. These commands are integral to the skill's primary function of finalizing character canonical identity and fact storage.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external files like CHARACTER_PROFILE.md, story architecture, and briefs, creating an attack surface where malicious content in these files could influence agent behavior.
Ingestion points: Workflow Step 1 identifies several ingestion points including CHARACTER_PROFILE.md, style rules, story architecture, and canon files.
Boundary markers: There are no boundary markers or explicit instructions to ignore embedded commands within the ingested data defined in the workflow.
Capability inventory: The skill includes the capability to write to local project files (01_story/characters.md and 00_project/canon.json) and execute Python scripts via the shell.
Sanitization: The workflow does not mention any sanitization, filtering, or validation steps for the external content before it is processed or written to canon files.

character-bible