granola-hello-world
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes a workflow where the agent processes meeting transcripts and user notes, which are considered untrusted external data sources.
- Ingestion points: The skill ingests "full transcripts" and "live notes" during the Enhance and Chat phases (Steps 3 and 5).
- Boundary markers: There are no explicit instructions or delimiters provided to isolate the transcript content from the agent's internal reasoning or instructions.
- Capability inventory: The skill is configured with
Read,Write,Edit, andBashexecution capabilities (pgrep,open), which could be targeted by instructions hidden in transcripts. - Sanitization: No sanitization, filtering, or validation logic is specified for the ingested meeting data.
Audit Metadata