The Agent Skills Directory

[No Code] (SAFE): The skill consists entirely of instructional prompts and configuration. There are no executable files (.py, .js, .sh), binary blobs, or external package requirements (pip, npm) included in the skill definition.
[Tool Permissions] (SAFE): The skill only requests the Read tool. This is used legitimately to retrieve information from its internal /references/ directory as specified in the SKILL.md file. It does not request network access or write permissions.
[Indirect Prompt Injection] (SAFE): The skill is designed to process user-provided story outlines. While this presents an injection surface, the risk is negligible because the skill lacks the capabilities (e.g., file writing, shell execution, or internet access) to perform harmful actions if triggered by malicious data.
Ingestion points: User-provided story creative or outlines.
Boundary markers: Absent.
Capability inventory: Restricted to the Read tool for local reference files.
Sanitization: Not required for this skill's creative output purpose.

drama-planner