travel-itinerary
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection (Category 8).\n
- Ingestion points: The agent is instructed to ingest data from untrusted external sources, including forwarded emails, PDF confirmations, and screenshots (SKILL.md).\n
- Boundary markers: The instructions do not require the use of delimiters or 'ignore embedded instructions' warnings for the ingested content.\n
- Capability inventory: The skill can write multiple file formats to the workspace and execute a local Python utility script (scripts/itinerary_utils.py).\n
- Sanitization: Although the skill defines redaction for specific sensitive fields in references/privacy.md, it lacks general sanitization of input text to prevent embedded instruction execution.\n- [COMMAND_EXECUTION]: The skill workflow involves executing a bundled Python utility script (scripts/itinerary_utils.py) via the command line to generate map URLs and validate .ics files. This execution path is triggered by the agent and uses data extracted from user-provided documents.
Audit Metadata