Claude Code and Codex Score Zero Running DeepSeek on kw-bench, OpenCode Gets 0.54

A new benchmark ran one model, DeepSeek v4.1 Flash, through four harnesses for two hours each. Two independent harnesses scored about half marks and the two lab harnesses scored zero. We go through that result and Arize's engineering behind agents, then Genie 4, the agent credential and exfiltration papers, two Anthropic stories on training data, and a few smaller items. kw-bench: a VAT reconstruction across three generations of SAP Arize on Code Mode Arize on how Alex is built Spine: a subtask-tree context runtime AgentBaton: mid-task model handoff DeepMind's Genie 4 Two Sigma on agent controls Keycard on agent credentials Log-To-Leak Stanford Daily on the Claude training-data allegation Latent Space interview with Anthropic bioRxiv: thirty agents, eighteen pipelines