Coverage tagged "GitHub".
Malware that takes technical cues from a poem to assemble a growing botnet by targeting open-source AI services has compromised more than 3,400 servers since April, Lumen Technologies’ Black Lotus…
Serial number: AV26-1007 Date: October 7, 2026 As of October 6, 2026, GitHub is affected by vulnerabilities in the following product: Enterprise Server 3.18.0 Prior to 3.18.16 3.19.0 Prior to 3.19.13…
Infostealers Are Actively Hunting AI Agents and Developer Keys – Warden Infostealer Initial announcement and feature breakdown of Warden Stealer on the Exploit.in dark web forum. A highly…
CVE-2026-21589: Critical Atlassian File Access Atlassian has disclosed CVE-2026-21589, a critical unauthenticated arbitrary file-access vulnerability affecting eight self-hosted products. The flaw…
Tuskira’s AI Agent Gateway is an open-source solution that sits between AI agents and everything they call: the MCP tool servers that connect them to services like GitHub and Jira, and the model…
arXiv:2610.06966v1 Announce Type: new Abstract: Indirect prompt injection (IPI) hides adversarial instructions in content that large language model (LLM) agents read at runtime. As agents compose…
arXiv:2610.07510v1 Announce Type: new Abstract: Developers can build LLM agents by adapting third-party models through benign post-training. We study a supply-chain threat in which an attacker…
arXiv:2610.07639v1 Announce Type: new Abstract: Coding agent harnesses mediate tool use and authorize actions, yet their security mechanisms and runtime effects remain incompletely characterized. We…
arXiv:2610.07645v1 Announce Type: new Abstract: Self-improving LLM agents increasingly distill successful experiences into persistent, reusable skills. Existing skill attack methods corrupt this…
arXiv:2610.07776v1 Announce Type: new Abstract: Website fingerprinting infers the websites visited by users from encrypted traffic metadata. However, models trained under fixed collection conditions…
Cryptographic Context Injection against a coding agent: one encrypted page makes GitHub Copilot CLI send local secrets to an attacker in 28 seconds.
Analysis of Blinder Tunnel, an Iran-nexus campaign using fake Dubai Airports recruitment lures and GitHub C2 malware to target critical infrastructure.
GitHub’s ReviewBench measures how well AI code review tools detect problems before software is released. Available in research preview through its website, the benchmark lets users compare review…
