ResearchSkillCraft tool composition benchmarkWweb·medium signalBenchmark for agent skill abstraction and cross-task reuse. Skill saving/reuse reduces token usage by 80%. Success rate correlates with tool composition ability. Code released. arXiv 2603.00718↳ Follow the threadPolicy dependency / Stack layerCodenotary ships AgentMon 3 for agent runtime policy and compliance loggingAI Agent StorePolicy dependency / Stack layerHalluSquatting weaponizes LLM hallucinations to push botnet malware into agentsThe Hacker NewsPolicy dependency / Stack layerMake context compaction a trained policy, not an inference-time heuristic — CompactionRL lifts SWE-bench up to +7 pointsarXiv 2607.05378Policy dependency / Stack layerReason Less, Verify More: Deterministic Gates Catch a Silent Policy-Violation Failure in Tool-Using AgentsarXivStack layer / Threat patternPick your agent sandbox primitive by threat model: Firecracker microVM vs gVisor vs V8 isolateNorthflankPolicy dependency / Stack layerPattern: A Provider-Abstraction / BYOK Layer Is Hardening Beneath Agent HarnessesGitHubStack layer / Threat patternVEXAIoT: arXiv paper demonstrates autonomous IoT vulnerability exploitation using AI agentsarXivStack layer / Threat patternGhostcommit hides prompt injection in PNG files to make AI coding agents exfiltrate repo secretsSecNews