🕵️ RESEARCH & DEEP DIVES
Srsly Risky Biz: China's botnets are worth disrupting — Risky Business News
ZhongBangKeJi CRMEB Custom Scheduled Task Feature save eval os command inject... — CVE ThreatInt
HDD Password Stored In PlaintextHDD password plaintext is stored in a UEFI va... — CVE ThreatInt
VariableEditSmm: Error checking of UEFI variables could cause buffer overflow... — CVE ThreatInt
SysPasswordDxe: Password hashes are exposed in runtime UEFI variables, leadin... — CVE ThreatInt
TOTOLINK CP450 cstecgi.cgi buffer overflowA vulnerability was found in TOTOLI... — CVE ThreatInt
HKUDS AI-Trader selfRegister API Endpoint routes_agent.py logic errorA vulner... — CVE ThreatInt
Context Inference Attacks Without Jailbreaks — arXiv cs.CR
Private Computation Space: Experience with Trusted Multi-Cluster Federated Learning for Agriculture — arXiv cs.CR
Ranked by the Matcher: A Reproducibility Audit of Knowledge Graph Extraction from Threat Reports — arXiv cs.CR
Skill-as-API: Confidential Multi-Agent Coordination for Agentic Software Engineering — arXiv cs.CR
Public-Sharing Labels and Verbatim Field Egress in an MCP-to-A2A Agent Configuration: A Controlled Multi-Model Study — arXiv cs.CR
Hearing the Whispers: Black-Box Membership Inference Attacks on Finetuned TTS Models — arXiv cs.CR
HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation — arXiv cs.CR
Towards Behavior Tree-Guided Vulnerability Detection with Lightweight LLMs — arXiv cs.CR
Agent Memory Is a Surface for Endogenous Authorization Laundering — arXiv cs.CR
Adversarial Vulnerabilities of Neural Biomarker Identification Systems — arXiv cs.CR
Agent Flight Recorder: Tamper-Evident Audit Trails with On-Chain Anchoring for Long-Horizon Tool-Using Agents — arXiv cs.CR
Bonded Recourse for Smart-Contract Settlement of Compensable Agent Side Effects — arXiv cs.CR
Privacy Amplification Without Independence: How Far Negative Dependence Carries the Guarantees of Poisson Subsampling — arXiv cs.CR
Pushing Forward Multi-Secret-Key Homomorphic Encryption for Private Average Aggregation — arXiv cs.CR
C$^2$T-OpenMax: A Novel Open-Set WiFi RF Fingerprinting Method via Center Constrained Learning and Confidence-Guided Tail Modeling — arXiv cs.CR
Implicit Manipulation for Skill Selection in LLM Agents with Semantic Matching — arXiv cs.CR
Type-Directed, Secure-by-Construction Enclave Partitioning for LLVM — arXiv cs.CR
Stored Is Not Supported: Typed Provenance and Assertion Guardrails for Persistent AI Agents — arXiv cs.CR
WeaveMark: Robust and Scalable Multi-bit LLM Watermarking via Coded Payload Spreading — arXiv cs.CR
Agentic Settlement Protocol: An Application Profile for Refundable, Delayed-Fulfilment Agent Commerce on Stablecoin Rails — arXiv cs.CR
Claude AI Now Controls Your macOS and Windows Computer in the Background — Cyber Security News