foresttrip-vacancy

Warn

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill uses npx -y @nomadamas/k-skill@0 to fetch instructions and update the skill. This pattern executes code from a remote NPM package that is not a verified trusted source.\n- [EXTERNAL_DOWNLOADS]: The skill requires the installation of Playwright and associated browser binaries. Playwright is a well-known service, so this is documented as part of the intended functionality.\n- [COMMAND_EXECUTION]: Instructions direct the user to run pip and npx commands to set up the environment and execute helper scripts.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data from the foresttrip.go.kr website, creating a potential surface for injection.\n
  • Ingestion points: Data is scraped from the forest recreation site in scripts/run_foresttrip_vacancy.py.\n
  • Boundary markers: The skill lacks explicit boundary markers for presenting scraped data to the agent.\n
  • Capability inventory: The script uses browser automation (playwright) and network requests (urllib.request).\n
  • Sanitization: The script filters for specific JSON keys and performs basic type validation for reservation counts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 1, 2026, 02:41 PM