The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads and processes external, untrusted content such as meeting transcripts and participant names. Malicious instructions embedded within a meeting's conversation or participant metadata could potentially attempt to influence the agent's behavior.
Ingestion points: Data enters the context via list_transcript_entries, get_transcript, and list_participants in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat these specific data fields as potentially untrusted content.
Capability inventory: The skill includes actions that can modify meeting state, such as update_space and end_conference, which could be abused if an injection is successful.
Sanitization: The documentation does not specify any sanitization, filtering, or validation performed on the retrieved text content.

google-meet