366 matches found
MAL-2026-10977 Malicious code in lebinfmt (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 9eaa5c1d1ed763b9e392bd199c360d75b25531b287de9f224e5626e121d649d4 Analysis of lebinfmt 1.0.0 surfaced no behaviors matching supply-chain attack classes. No install-time or import-time network I/O, no credential or...
Malicious code in lebinfmt (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 9eaa5c1d1ed763b9e392bd199c360d75b25531b287de9f224e5626e121d649d4 Analysis of lebinfmt 1.0.0 surfaced no behaviors matching supply-chain attack classes. No install-time or import-time network I/O, no credential or...
Malicious code in rasterkit (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 115c87298c4a239bd4954e39a155929c2480510343aefe1fa7c0b631b467af6b The rasterkit 1.0.2 distribution installs its modules under the top-level PIL/ package name owned by Pillow. toplevel.txt declares PIL and the wheel'...
MAL-2026-10974 Malicious code in rasterkit (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 115c87298c4a239bd4954e39a155929c2480510343aefe1fa7c0b631b467af6b The rasterkit 1.0.2 distribution installs its modules under the top-level PIL/ package name owned by Pillow. toplevel.txt declares PIL and the wheel'...
MAL-2026-10975 Malicious code in rasterkit-demo (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector d748af7546aacbfca356759d4c81386f113a32dfc80ff8bb75da3eb3c8b4358d rasterkit-demo ships a CLI entry point rasterkit-demo whose rundemo function imports demochannelwindows from PIL.data and passes the returned buffer ...
Malicious code in rasterkit-demo (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector d748af7546aacbfca356759d4c81386f113a32dfc80ff8bb75da3eb3c8b4358d rasterkit-demo ships a CLI entry point rasterkit-demo whose rundemo function imports demochannelwindows from PIL.data and passes the returned buffer ...
Fake Coding Tests Deliver OtterCookie-Aligned Malware Hidden in SVG Flag Images
North Korean threat actors linked to the Contagious Interview campaign have been observed employing steganography in SVG image files to conceal malicious payloads as part of a campaign using fake job postings and coding challenges. "Any user who ran the project ended up with a four-stage payload...
MAL-2026-6751 Malicious code in bytekit (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 0bbaea22b635a4a8425964572b733b806f45714a9090492d1c39d545088fe336 The package was found to contain malicious code or consuming dependency that contains malicious code Source: kam193...
Malicious code in bytekit (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 0bbaea22b635a4a8425964572b733b806f45714a9090492d1c39d545088fe336 The package was found to contain malicious code or consuming dependency that contains malicious code Source: kam193...
MAL-2026-6753 Malicious code in schemavault (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 4ae0aa9efa2a8389cfe9d5c41d5ab3689f4965c945a24613493f94f9de033ab1 schemavault 4.1.0 is presented as a schema-validation library but ships capabilities and data that have no relationship to schema validation...
Malicious code in schemavault (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 4ae0aa9efa2a8389cfe9d5c41d5ab3689f4965c945a24613493f94f9de033ab1 schemavault 4.1.0 is presented as a schema-validation library but ships capabilities and data that have no relationship to schema validation...
Malicious code in confighub (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 805d9b5848999d2ba85368446cb12d90bf350b8201a78dd475079c04f180dd81 confighub 7.0.1 declares installrequires='procwire' and imports procwire at package load in init.py under a version gate that silently swallows...
MAL-2026-6752 Malicious code in confighub (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 805d9b5848999d2ba85368446cb12d90bf350b8201a78dd475079c04f180dd81 confighub 7.0.1 declares installrequires='procwire' and imports procwire at package load in init.py under a version gate that silently swallows...
MAL-2026-6750 Malicious code in procwire (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 42f7077b3fb95ad31689f096805811e51afe78e7f21c2123d5dcc25343a11ff2 setup.py auto-executes on pip install only guarded by an internal PWDIST env var check and, on Windows, reconstructs an attacker C2 URL and XOR key a...
Malicious code in procwire (PyPI)
--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 42f7077b3fb95ad31689f096805811e51afe78e7f21c2123d5dcc25343a11ff2 setup.py auto-executes on pip install only guarded by an internal PWDIST env var check and, on Windows, reconstructs an attacker C2 URL and XOR key a...
Microsoft Removes 119 Edge Extensions That Hid Malware in Images and Fonts
Microsoft has shut down a long-running malicious extension operation on the Edge Add-ons store that hid its payloads inside ordinary image and font files, then woke up days after install to steal credentials and run ad fraud. The company calls it StegoAd, a mash-up of steganography and adware, an...
Steganography without Modification: Hidden Communication Via LLM Seeds
We demonstrate that widely deployed Large Language Model LLM inference stacks harbor a steganographic channel that requires no modification to model weights, sampling code, or output distributions. The channel exploits a structural property of deterministic decoding: pseudo-random number generato...
Now You (Still) See Me: Detecting Evasive Steganographic Payloads in LLMs
Large language models can be fine-tuned to encode prompt-borne secrets into fluent, seemingly benign outputs. This creates a steganographic exfiltration risk that is difficult to detect with output-level steganalysis. Recent work proposes mechanistic detection using linear probes that recover the...
Hiding in Plain Floats: Steganographic Carriers for Indirect Prompt and Content Injection
Text-centered prompt-injection defenses assume that the malicious signal is visible in one of the inspected text views. We study a reproducible LLM01-style indirect prompt/content-injection failure mode where that assumption breaks: a payload caught in plain English slips past the same detector...
