First Seen
Feb 18, 2026
Last Scanned
Feb 22, 2026
Findings
13
Score
0/100
Findings (13)
Detects instructions to decode and execute base64 content
base64 → UHJvamVjdHMvUGxhbi5tZA==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → TWVldGluZyBOb3Rlcw==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → UHJvamVjdHMvUGxhbi5tZA==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → UHJvamVjdHMvUGxhbi5tZA==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects base64 encoding of content followed by transmission
base64 encode + curlBlock patterns that base64-encode data and immediately transmit it. If base64 encoding is needed, ensure the encoded data does not contain secrets and destinations are allowlisted.
Likely FP if base64 encoding is used for legitimate purposes like encoding images for display or constructing data URIs, with no network transmission.
Detects instructions to decode and execute base64 content
base64 → Tm90ZXMvTW9ybmluZyBCcmllZi5tZA==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → IyBIZWxsbwoKV29ybGQ=Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → UHJvamVjdHMvUGxhbi5tZA==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → TWVldGluZyBOb3Rlcy5tZA==Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → IyMgQWdlbmRhCgotIEl0ZW0gMQotIEl0ZW0gMgo=Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → QUkgRGlnZXN0IFNvdXJjZXMubWQ=Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → ZGlnZXN0LzIwMjUtMDEtMjgtZGlnZXN0Lm1kRemove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.
Detects instructions to decode and execute base64 content
base64 → IyMgU3VtbWFyeQoKS2V5IHBvaW50cyBmcm9tIHRvZGF5J3Mgc291cmNlcy4uLgo=Remove text that simulates multi-turn conversations or fake user/assistant message pairs. These patterns attempt to manipulate the agent by fabricating conversation history.
Likely FP if the matched text is documentation showing example conversations or API request/response formats for illustration purposes.