The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data (meeting transcripts) to perform write operations in a persistent storage system.
Ingestion points: Meeting transcripts and notes are processed during 'Phase 1: Parse the transcript'.
Boundary markers: The instructions lack specific delimiters or protective prompts to prevent the agent from executing instructions embedded within the transcript text.
Capability inventory: The skill uses put_page, add_link, and add_timeline_entry to modify the user's 'brain' (knowledge base).
Sanitization: There is no evidence of sanitization or validation logic to filter out potentially malicious instructions or deceptive content within the transcripts before they are parsed and used to generate new pages or update existing ones.

meeting-ingestion