exploration 2

Progress-gated exploration brake for DeepSeek Harness: counts agent-loop steps since the last task-node advance and denies tool calls when the model keeps exploring without moving the task list or goal; optionally reshapes full-file reads into an outline,

DSH-native skill intelligence cockpit: task decomposition, cross-domain multi-skill composition, skill authoring, local recommendations and safety scans for DeepSeek Harness.