brainstorming

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

• Third-party content exposure detected (high risk: 0.70). The skill is sourced from an upstream public GitHub repository (“Upstream: https://github.com/obra/superpowers”), and the runtime workflow likely ingests that repository’s README/assets as free text into the agent’s LLM context, which is outsider-authored public web content.