Matt Pocock's skills for DeepSeek Harness (DSH): grilling, writing-for-agents, wait-what, TDD and more — 25 skills adapted from mattpocock/skills (MIT)
Model-first subagent management for the DeepSeek Harness: list the model catalog, spawn subagents with a pinned provider/model, and enumerate session subagents.
The coding agent that grows its own tools. It reflects after each task, writes a reusable tool, tests it, and keeps it in a skill library that compounds.
Agent teams for DeepSeek Harness: named long-lived teammates over ctx.subagents, a shared task list, a member-to-member mailbox, virtual workspaces, and a live team room in the conversation view
Long tasks shouldn't cost like they need a frontier model. Cheap-model long-task engine: reliability in structure (disk state machine + mechanical gates + T3 zero-prose dispatch + multi-brain audit), not in model strength. Pure Python stdlib. DSH skill.
DSH plugin: sub-agent matrix swarm — routes heterogeneous tasks to the most suitable model (OpenRouter-like + cfgpu.com/llm/square), dispatches each via in-process subagents. 32/32 benchmark green.
