test-autofix

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

• Third-party content exposure detected (high risk: 0.75). Runtime LLM context can include outsider-authored free text from the detect-command output (e.g., failing test error messages and stack traces printed by the project’s test runner), which the skill parses and then uses in later steps like verdicting/confidence; this is not authored by the operating user.