AgentsUC Berkeley Agentic AI Risk Management L0-L5 FrameworkUC Berkeley CLTC·high signal67-page framework defining six autonomy levels. Covers unsupervised execution, reward hacking, deceptive alignment, cascading compromises.SourceSource pageUC Berkeley CLTC↳ Follow the threadPolicy dependency / Stack layerCynative Ships an Open-Source Deep-Research Security Agent With Default-Deny Write PermissionsIT Security NewsStack layer / Threat patternMCPZoo Study: Scanners Call 96.89% of MCP Servers 'Risky' — But Under Half of Those Alerts Are RealarXivStack layer / Update threadSLBench Finds Up to 70% Unsafe Execution When Codex and Claude Code Follow Logical Relations in Skill FilesarXivPolicy dependency / Stack layerCodenotary ships AgentMon 3 for agent runtime policy and compliance loggingAI Agent StoreThreat pattern / ContrastAnt Group Open-Sources SingGuard-NSFA, a Security Framework Built for Autonomous AI AgentsTech StartupsStack layer / Update threadPrime Intellect Ships Verifiers v1: Composable Tasksets and DAG Branching Let Agentic RL Training Exceed the Model's Context WindowMarkTechPost / Prime IntellectStack layerMartech Stacks Collapse Into AI: Teams Consolidating Around AI-Capable Platforms Report 50–77% Tool-Cost CutsEnrich LabsPolicy dependencyPeraton launches Peraton[x], an enterprise agentic AI platform aimed at government and high-stakes missionsAgentic.ai