/
initial_access_vector:supply_chain_compromise × victim_sectors:technology × criticity:medium × limpar

Resultados

3 casos · pág. 1/1

f5.com · medium

F5 Labs: Chinese open-weight AI models show strong cyber capability but uneven resistance to prompt injection and jailbreaks

F5 Labs assesses the security risks and rewards of Chinese open-weight LLMs (Qwen, GLM, Kimi, MiMo, MiniMax, DeepSeek). It finds that strong offensive and defensive cyber capability does not translate into consistent resistance to attack. Cited examples include Kimi K3 finding 19 vulnerabilities in Redis and GLM 5.2 being self-hosted by H…

safedep.io · medium

Cursor's VS Code import installs latest Open VSX extension versions, exposing agentic IDE users to supply chain risk

SafeDep research shows that Cursor's Import VS Code Configuration step copies only the publisher.extensionname of each installed extension and drops the version. Cursor then installs whatever version Open VSX (run by the Eclipse Foundation) calls latest, through its proxy marketplace.cursorapi.com. In SafeDep's test, three extensions pinn…

safedep.io · medium

Researcher squats 21 npm names matching bin entries of Google-scoped packages, using postinstall fingerprint beacons

On August 12, 2026, an npm user (rootdaddy-msrc, author field r00tdaddy) published 21 packages that squatted the unscoped CLI binary names declared in the bin field of Google-affiliated scoped packages, such as ngsw-config (from @angular/service-worker) and bazelisk (from @bazel/bazelisk). Scoped publishing, registry allowlists and lockfi…