swift-concurrency

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md "Fetching More Docs" workflow explicitly instructs the agent to fetch documentation from a public third-party Markdown mirror (e.g., mapping /documentation/swift/task to https://sosumi.ai/documentation/swift/task), which means the agent will ingest and act on open web content that could contain untrusted instructions.