obc

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

• Third-party content exposure detected (high risk: 0.80). The skill's command list and docs explicitly include /xdaily ("Pulls X/Twitter posts into daily notes") in references/ObsidianCommands.md and SKILL.md, showing the agent ingests public, user-generated social media content and reads/interprets it as part of workflows — which could allow indirect prompt injection.