cloudbase

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Standalone skill fallback" in SKILL.md explicitly instructs the agent to fetch and read public SKILL.md and sibling files from a public raw Git URL (https://cnb.cool/.../SKILL.md), and those external documents are required reading that directly influence routing, tool use, and next actions—exposing the agent to untrusted third‑party content that could inject instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to fetch and read remote skill files from https://cnb.cool/tencent/cloud/cloudbase/cloudbase-skills/-/git/raw/main/skills/cloudbase/SKILL.md (which would directly control agent routing/prompts), and also requires running npx @cloudbase/cloudbase-mcp@latest (which fetches and executes remote code) — both are runtime external dependencies that control instructions or execute code.