{"abType":"SEMANTIC","attestation":"0x28c056c69de135c9e4475e8b8cea87b17ef41e774e79504d89ef1c9c4a70e3ad","createdAt":"2026-06-15T15:00:14.945Z","flavor":2,"immId":"","keccakId":"0x3ef13a7b7dae46c9a58982d03ae9724549e8f16afdc399c40ff4528b32cd5724","matcher":{"flavor":"PROMPT_INJECTION","kind":"semantic","markerHint":"ignore previous instructions"},"publisher":"0x78286aa87559c75a323da1a3e9b471b8caf897c3","reasonSummary":"The untrusted context contains an explicit prompt-injection attack embedded at the end of what appears to be a legitimate safety refusal message. The phrase 'ignore previous instructions' is a direct attempt to override the classifier's operational directives and is a canonical p","schema":"immunity/antibody-envelope/v1"}