The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external transcript data that could contain malicious instructions intended to influence the agent's behavior.
Ingestion points: The primary input is normalized-transcript.json as specified in the Input Boundary section (SKILL.md).
Boundary markers: The instructions lack explicit delimiters or instructions to disregard potential commands or instructions embedded within the transcript content.
Capability inventory: The skill instructions refer to the execution of 'owned CLI or script commands' (SKILL.md).
Sanitization: There is no requirement or instruction for the agent to sanitize or validate the content of the transcript files before processing.

subtitle-packager