lovebug-report

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

• Third-party content exposure detected (low risk: 0.10). Runtime workflow ingests outsider-authored free text when it performs searchLovebugRegions using an attacker-controlled query (e.g., user text like “중랑”), then uses the results from https://xn--2i0bt2q2wd1wb.com/api/map/gu-score and related JSON endpoints to answer.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL explicitly requires running "npx -y @nomadamas/k-skill@0 instruct lovebug-report" which fetches and executes a remote npm package at runtime to provide the primary instructions, so remote code controls the agent's prompts.