norikae-guide

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md and README explicitly instruct the agent to build Yahoo! 乗換案内 URLs and fetch live result pages from the public site https://transit.yahoo.co.jp (using scripts/fetch_norikae_routes.py), so untrusted third‑party webpage content is ingested and parsed for route decisions and summaries.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill's runtime fetch script (scripts/fetch_norikae_routes.py) retrieves live pages from https://transit.yahoo.co.jp/search/result and injects extracted HTML/text into agent responses, so external content fetched at runtime can directly influence what the agent outputs.