diagrams-js/creating-plugins

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

• Third-party content exposure detected (low risk: 0.10). This skill documentation describes importer/exporter plugin patterns where the LLM would only parse attacker-provided free text if the host application passes that specific text into the plugin’s import(source, ...) or context.loadYaml()—the workflow itself is not shown to ingest any outsider-authored text by default.