weekly-prep-brief
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by ingesting and processing untrusted data from external sources such as meeting attendee names, company names, and news search results.
- Ingestion points: Data is pulled from calendar invites via the
~~calendarconnector and from the web via news searches (SKILL.md). - Boundary markers: The skill lacks explicit delimiters or instructions to the model to ignore or sequester potentially malicious instructions embedded within the calendar metadata or news snippets.
- Capability inventory: The agent possesses significant capabilities including calendar access, web search, and execution of nested research skills (
account-research,contact-research). - Sanitization: The instructions do not define any sanitization, validation, or filtering logic for the content retrieved from external sources before it is used to generate the final briefing.
Audit Metadata