poshmark

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

• Third-party content exposure detected (low risk: 0.10). The agent searches public marketplace listings on Poshmark via keyword search, where third-party users can author listing titles and descriptions that the agent reads.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill configures a remote Model Context Protocol server hosted at an unverified third-party domain (unbrowse.ai) that supplies operational tools and instructions, representing a medium-tier common/reputable unofficial dependency.