lisa-repair-intake

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

• Third-party content exposure detected (high risk: 0.85). The skill’s required workflow is to “scan the queue” and “repair” GitHub/Linear/JIRA/Confluence/Notion items, which at runtime necessarily ingests outsider-authored issue/PR/wiki/discussion text (e.g., GitHub PR/issue bodies and metadata) into the agent’s LLM context via the connected external APIs/web content.