openharmony-ability-kit

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

• Third-party content exposure detected (high risk: 0.70). The documentation includes APIs that open or start abilities with arbitrary external URLs (e.g., this.context.openLink('https://example.com') and Want.uri used for implicit starts), which clearly allows the agent to open/display untrusted public web content as part of its runtime workflow.