The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing external data through its multi-agent production pipeline.
Ingestion points: All user-provided topics, keywords, and files are aggregated into _workspace/00_input.md which serves as the primary context for the team.
Boundary markers: There are no instructions to use delimiters or ignore instructions within the processed files to prevent accidental execution of embedded commands.
Capability inventory: The agents have the capability to create and modify files in the local workspace and communicate via structured messaging.
Sanitization: The pipeline does not specify any sanitization, filtering, or validation of user-provided content before it is passed to agents.

podcast-studio