modernize-flp-sandbox

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

• Third-party content exposure detected (medium risk: 0.65). Runtime LLM context can be influenced by outsider-authored free text because the skill reads and parses HTML/JS files under the app’s webapp/test/ (including OPA test pages and any externalized externalUshellConfigFile contents) and then emits Gap-Report text that may incorporate extracted snippets/values from those files into the agent’s LLM prompt; these files are not authored by the operating user by definition.