ai-execute-task

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

• Third-party content exposure detected (medium risk: 0.30). Required runtime workflow reads outsider-authored free text from the user’s Obsidian vault by loading the PRD/tech spec/tasks and the specific NN-task.md specified by the feature and task selection (Read/ls from $OBSIDIAN_AI_VAULT/engineering/<project>/workplans/<feature>/...).