The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read existing project files and search the web for context, which creates an attack surface for indirect prompt injection if those sources contain malicious instructions. * Ingestion points: The skill reads scripts, treatments, character notes, outlines, and web search results into the agent context as part of its core workflow. * Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore embedded commands within the ingested data. * Capability inventory: The skill utilizes web search, filesystem read, and filesystem write capabilities (specifically writing markdown files to the project directory). * Sanitization: There is no evidence of content sanitization or validation for the data retrieved from external project files or search results.

alterlab-cdm-screenwriter