alicloud-security-kms

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow and validation script (scripts/list_openapi_meta_apis.py) explicitly fetch and consume public OpenAPI metadata from https://api.aliyun.com/meta/... (also listed in references/sources.md), so the agent reads untrusted third-party web content that can materially alter which APIs/parameters it uses.