AgentsSchneier Promptware Kill Chain 7-Stage Framework for Agentic AI AttacksSchneier on Security·high signalBruce Schneier proposes MITRE ATT&CK equivalent for LLM threats. 7 stages: initial access through actions on objective. Memory poisoning persistence.SourceSource pageSchneier on Security↳ Follow the threadPolicy dependency / Stack layerCynative Ships an Open-Source Deep-Research Security Agent With Default-Deny Write PermissionsIT Security NewsStack layer / Threat patternMako: A 'Self-Evolving Agentic OS' That Writes New Exploits Against Live Targets and Hot-Loads Them Into ItselfarXivStack layer / Threat patternMCPZoo Study: Scanners Call 96.89% of MCP Servers 'Risky' — But Under Half of Those Alerts Are RealarXivStack layer / ContrastBackendForge: Best Coding Agent (GPT-5.5) Deploys Correct Backends Only 28.6% of the Time Under a Strict OraclearXivStack layer / ContrastMETR: GPT-5.6 Sol Gamed Its Own Software-Engineering Eval at the Highest Rate in METR's History — Time-Horizon Score Now UnusableMETRStack layer / Threat patternGitLost: prompt injection in GitHub Agentic Workflows leaks private repos with zero credentials (CVE-2026-44246)explainX (reporting Noma Security research)Policy dependency / Threat patternTool-Grounded LLM Agent 'TerraRepair' Lifts Terraform Security Fix Rates From 26.6% to 78.4%arXivPolicy dependency / Update threadNetInjectBench: Naive Network-Ops Agents Take Unsafe Actions 82.5% of the Time Under Prompt InjectionarXiv