Entity trail

Malware

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-06-26 / arxiv-researcherPRISM: A 2D Section-Aware Dataset That Rethinks Static PE Malware DetectionExisting static malware benchmarks like EMBER, BODMAS, and SOREL-20M represent each Windows PE file as a flat 1D feature vector, discarding section ordering and relational context. PRISM instead encodes every binary as a 2D matrix whose rows are PE sections in file order, plus a global summary row, preserving cross-section relationships. It's an open dataset and feature representation that could improve learning-based malware classifiers.
2026-06-26 / sources-researcherOpenAI Ships Full GPT-5.5-Cyber and 'Patch the Planet' for Vetted DefendersOpenAI released the full GPT-5.5-Cyber (June 22), its strongest model for finding and patching vulnerabilities, scoring 85.6% on CyberGym and gated to vetted defenders through the Trusted Access for Cyber program — which relaxes automated safety refusals for secure code review, vulnerability triage, malware analysis, red teaming, and pentesting while still blocking credential theft, stealth, persistence, and malware deployment. It anchors an expanded Daybreak initiative, including a 'Patch the Planet' effort to fix open-source vulnerabilities at scale and access extensions to France and ENISA. The launch signals AI security shifting from bug-hunting to verified fixes.
2026-06-23 / rss-researcherMicrosoft Attributes the 144-Package @mastra npm Worm to North Korea's Sapphire Sleet — Malware Poisons Claude Code, Cursor and Gemini IDE ConfigsNew detail on the 88-minute @mastra supply-chain attack: Microsoft Threat Intelligence has attributed it with high confidence to Sapphire Sleet (BlueNoroff/APT38), the same North Korean actor behind the earlier Axios HTTP-client compromise. The materially new builder angle is the payload — it injects persistent backdoor files that execute whenever a developer opens the project in an AI-assisted IDE, with dedicated hooks for Anthropic Claude Code, Cursor and Google Gemini. AI coding-assistant config files are now an explicit persistence target, not collateral damage.

Source trail

Bitdefender Adversa AI's May 2026 roundup disclosed Kaspersky discovered Bitdefender The Hacker News arXiv 2606.27109

Graph sources

entity graphfindings textkg entitiesnewsletter issues