The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local script (scripts/query.mjs) using the Node.js runtime to process train ticket data and generate HTML reports.
[INDIRECT_PROMPT_INJECTION]: The skill takes user-provided destination and origin information and uses it to construct shell commands, creating a potential command injection surface.
Ingestion points: User-supplied travel preferences including destination, departure city, and dates gathered in Step 1 and used in Step 3 of the SKILL.md workflow.
Boundary markers: Absent. The instructions provide a direct command template (node [SKILL_DIR]/scripts/query.mjs [from] [to] ...) without delimiters to isolate untrusted user data.
Capability inventory: The skill possesses the ability to execute local scripts and write to the filesystem (e.g., caching station data in data/stations.json and saving HTML output).
Sanitization: Absent. There are no prompts instructing the agent to validate or escape city names before passing them as command-line arguments.
[EXTERNAL_DOWNLOADS]: The included scripts fetch live station metadata and ticket information from official 12306.cn domains, which is a well-known service for national rail transport.

travel-planner