coral-create-source-spec

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to "Read the provider API docs" and to "validate by querying" external HTTP-backed providers (e.g., use coral source test and "fetch actual rows"), which requires ingesting untrusted public API/docs and runtime responses that can influence subsequent spec-generation and tool actions.