grey-haven-creative-writing

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

• Third-party content exposure detected (high risk: 0.80). The skill's pipeline explicitly instructs agents to read and analyze external competitive URLs and source libraries (see the project config "Competitive URLs" table and Phase 1/Phase 2 prompts: "Read the project config... competitive URLs" and "Source library" / "Source everything"), so the agent will ingest and interpret untrusted public web content as part of its workflow.