local-recurring-activity-planning
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted information gathered via the web and browser tools, including venue pricing, regulations, and FAQs. This exposure allows for potential attacks where a malicious website provides instructions to bias the agent or manipulate the generated outreach message.\n
- Ingestion points: Sections 4 and 4b of SKILL.md require the agent to visit and verify details from third-party websites and social profiles.\n
- Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted web content from the agent's core instructions during processing.\n
- Capability inventory: The skill utilizes web, browser, calendar, and memory tools.\n
- Sanitization: There is no instruction to sanitize or escape data retrieved from the web before it is included in recommendations or the outreach message template.
Audit Metadata