ktx-booking
Warn
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
korail2-ncardandpycryptodomePython packages from external registries. Whilepycryptodomeis a standard library for cryptographic operations,korail2-ncardis a specialized package for Korail interactions that is not from a recognized trusted vendor. - [COMMAND_EXECUTION]: The skill relies on executing a local script
scripts/ktx_booking.pyto perform train searches, reservations, and cancellations. This script is the primary execution logic for the skill's capabilities. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted user input (e.g., station names, travel dates, and train IDs) and processes this data through tools with command execution and file-system access capabilities.
- Ingestion points: User-provided inputs for departure and arrival stations, dates, and times defined in
SKILL.md. - Boundary markers: The skill instructions do not specify any delimiters or safety markers to isolate user-provided data from command parameters.
- Capability inventory: The skill can execute shell commands via
python3andpip, and it manages a local credential file at~/.config/k-skill/secrets.env. - Sanitization: No sanitization or validation logic is described to filter or escape user-provided arguments before they are passed to the helper script.
- [DYNAMIC_EXECUTION]: The skill uses a
python3 -cinline command to dynamically check for the presence of required libraries (korail2andCrypto) before attempting to run the main script.
Audit Metadata