travel-planner
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill demonstrates safe behavior by requiring user confirmation of budgets and preferences before performing any planning actions. It enforces strict data provenance rules to ensure the reliability of information provided to the user.
- [INDIRECT_PROMPT_INJECTION]: The skill presents a potential attack surface by ingesting external web content through its research requirements.
- Ingestion points: Travel-related data is collected from the internet via search and fetch tools during Step 2 of the process.
- Boundary markers: The prompt mandates clear separation of official facts from community opinions and requires the agent to explicitly cite sources and retrieval dates.
- Capability inventory: No high-risk capabilities such as file system writes, privilege escalation, or arbitrary shell command execution are utilized by the skill.
- Sanitization: The instructions require the agent to cross-reference multiple sources for reputation-based information and strictly prioritize official government or venue websites for factual data.
Audit Metadata