events
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional documentation and strategic planning frameworks. It contains no executable scripts, shell commands, or binary files that could lead to unauthorized system actions.
- [SAFE]: A dedicated 'Permission Boundary' section explicitly requires the agent to obtain user authorization before performing any high-impact actions, such as purchasing sponsorships, travel, registering attendees, or writing to CRM systems.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies external data sources (attendee lists, web scraping, and prospecting data) as potential attack surfaces and proactively instructs the agent to treat this data as 'untrusted evidence.' The 'Unified Skill Contract' provides clear guidance on handling untrusted input, effectively mitigating the risk of indirect prompt injection.
- [SAFE]: The 'Connector & Tool Layer' documentation enforces security best practices by explicitly prohibiting the storage of API keys in skills or source control and defaulting integrations to read-only/dry-run behavior.
Audit Metadata