PRACTICE / AGENT LAB
Проверяем локальный firewall от prompt injection для coding-агента: Проверяем локальный firewall от prompt injection для coding-агента
This bounded field note explains может ли вредоносный текст в выводе инструмента незаметно изменить действия ии-агента. and defines a reproducible evaluation of Проверяем локальный firewall от prompt injection для coding-агента without claiming unverified production results.
Test boundary
The test addresses может ли вредоносный текст в выводе инструмента незаметно изменить действия ии-агента.. It is limited to the stated scenario and does not claim production reliability.
Minimal scenario
Define one repeatable test, keep the input and model settings stable, and record each run with a stable identifier. The expected result is: Воспроизводимый набор атак и таблица того, какие инъекции Bouclier AI блокирует или пропускает..
Verification
Run the same cases several times, save the measured outputs and compare the result against the acceptance criteria. Do not replace measurements with a model-generated conclusion.
Limitations
This is a reproducible field test, not a security certification or a guarantee of production behavior.
Related measurements
This material uses the measured query cluster: . It does not promise a search result.
Browse practical guides · Open the glossary