kuroco-admin-api

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 (SKILL.md) instructs the agent to call the target Kuroco management instance (e.g., via kuroco-admin discover, help, advise, and exec against the provided management URL) and to ingest API responses and site content (modules, schemas, topics/members data) which the agent must read and use to decide endpoints/parameters and next actions, meaning untrusted/user-provided content from the third‑party site can materially influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill invokes the management endpoint (e.g., https://example.g.kuroco-mng.app) at runtime (login/exec/advise/discover/help) and the CLI's advise/discover responses include AI-generated "steps"/"body_example" that the agent uses to determine subsequent actions, so remote content from that URL can directly control agent instructions.