morning-briefing

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests and processes untrusted data from external sources to generate the briefing.
  • Ingestion points: Fetches unread emails via Gmail (Step 3), calendar events via Google Calendar (Step 2), and news via web search (Step 4).
  • Boundary markers: Absent. The briefing composition instructions in Step 5 do not provide delimiters or 'ignore' instructions to the model regarding content found within the fetched data.
  • Capability inventory: The skill has the capability to send emails and SMS messages to the user and to execute a local Python script for database queries.
  • Sanitization: Absent. The skill does not perform any validation or sanitization on the text retrieved from emails or search results before including it in the final delivery.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 02:09 PM