reflect

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly downloads and merges defaults from an external "marketplace" (see "specweave refresh-marketplace" / "Step 1: Download latest marketplace" and linked GitHub commit) and then reads/interprets those marketplace-provided learnings into skill MEMORY.md files, exposing the agent to untrusted third-party content.