travel-planning
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads and installs CLI tools from the vendor's repository (better-world-ai/x-cli) and a third-party Homebrew tap (xpzouying/agent-cli). While the vendor's own resources are expected, the third-party tap represents an external dependency.\n- [COMMAND_EXECUTION]: Setup and operation require the agent to execute various shell commands, including environment configuration (brew tap, brew install), source code management (git clone), and binary compilation (go build).\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from external travel platforms (Ctrip, Booking.com, and Xiaohongshu), which could potentially contain adversarial instructions aimed at influencing the agent's behavior.\n
- Ingestion points: Data retrieved from external websites via the outputs of ctrip-cli, booking-cli, and xiaohongshu-cli.\n
- Boundary markers: None identified in the skill instructions to separate external data from system prompts.\n
- Capability inventory: The skill has capabilities for shell command execution and local file system interaction.\n
- Sanitization: No sanitization or validation logic is specified for the external data before it is processed by the agent.
Audit Metadata