hotel-travel-leads
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process data from external booking platforms, tourist board registers, and association directories (SKILL.md). This creates a potential ingestion surface for indirect prompt injection. However, the risk is inherent to the primary function of lead generation. The skill does not provide automated scripts or high-privilege commands that would allow for exploitation without user oversight. (1) Ingestion points: External booking websites and licensing registers; (2) Boundary markers: Absent; (3) Capability inventory: Standard agent text processing and the listed MCP tool; (4) Sanitization: None identified.
- [EXTERNAL_DOWNLOADS]: The skill contains a setup command for a Model Context Protocol tool hosted at
https://mcp.manypi.com/mcp(SKILL.md). This domain is owned by the skill's author (manypicom) and the tool is presented as a legitimate extension for lead generation tasks.
Audit Metadata