knowledge-ops

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

• Third-party content exposure detected (medium risk: 0.30). All runtime “input” text for the required workflows comes from either (a) a user-supplied local inventory JSON (--input) or (b) a user-supplied directory of local markdown files (--root) whose YAML frontmatter/fields are read by the scripts, so outsider-authored free text can be ingested whenever the user provides that data.