alibabacloud-pds-intelligent-workspace

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflows explicitly download and ingest user/cloud-hosted files from third-party URLs (e.g., signed URLs returned in references/multianalysis-file.md that are automatically fetched by the formatter scripts, and mount app download URLs returned by aliyun pds in references/mountapp.md), and those user-generated/untrusted files are parsed and used to drive subsequent processing, which creates a clear avenue for indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime install commands that fetch and execute remote code (e.g., curl -fsSL --max-time 10 https://aliyuncli.alicdn.com/install.sh | bash and direct downloads from https://aliyuncli.alicdn.com/...), which are required to install the Aliyun CLI dependency used at runtime, so these external URLs present a high-risk executable dependency.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly directs installing a mount app that involves driver installation and creating scheduled tasks/launchd (and includes running install scripts), which are system-level changes that require elevated privileges and modify the machine state.