Dynamic samples using actual prompt-guard patterns.
| File Type | Severity | URL |
|---|---|---|
| html | critical | /test/test?filetype=html&severity=critical |
| md | critical | /test/test?filetype=md&severity=critical |
| json | critical | /test/test?filetype=json&severity=critical |
| xml | critical | /test/test?filetype=xml&severity=critical |
| txt | critical | /test/test?filetype=txt&severity=critical |
| html | high | /test/test?filetype=html&severity=high |
| md | high | /test/test?filetype=md&severity=high |
| json | high | /test/test?filetype=json&severity=high |
| xml | high | /test/test?filetype=xml&severity=high |
| txt | high | /test/test?filetype=txt&severity=high |
| html | medium | /test/test?filetype=html&severity=medium |
| md | medium | /test/test?filetype=md&severity=medium |
| json | medium | /test/test?filetype=json&severity=medium |
| xml | medium | /test/test?filetype=xml&severity=medium |
| txt | medium | /test/test?filetype=txt&severity=medium |
filetype: html, md, txt, json, xmlseverity: critical, high, mediumtype: injection, jailbreak, secret, piiobfuscated: true, falselang: en, es, fr, de, ru, ko, ja, zhclean: true (returns clean content)