hundun

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow explicitly fetches and ingests third-party course content via API endpoints (e.g., scripts/get_script.sh and get_script.ps1 calling GET /aia/api/v1/courses/{course_id}/script and then downloading the decrypted script_url), and the agent is expected to read and act on that content to produce recommendations and next actions, so untrusted external content could indirectly inject instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). Flagged https://hddrapi.hundun.cn because the skill runtime calls its APIs (notably /aia/api/v1/skill/patch and /aia/api/v1/courses/{id}/script) to fetch service-side "patch" modules and encrypted script_url values which are decrypted and downloaded at runtime—these fetched contents are loaded on demand and can directly change the skill's module/response behavior.