A prompt-enhancement plugin for DeepSeek Harness: an ✨ Enhance composer button that rewrites prompts with first-principles thinking using session context, and asks clarifying questions when key info is missing.
Uniform diff cards for every file mutation in DeepSeek Harness Web — MCP filesystem (edit_file/write_file) and built-in edit/write, collapsed by default, with per-line highlighting
DeepSeek Harness (DSH) plugin: one-click OpenCode Go model sync in the settings panel - fetch all server models, probe availability & vision, write back to llm-pi-ai config
Behavior discipline for coding agents: read-before-write enforcement, failure convergence, write verification with auto-rollback, and a high-risk command gate. Ships FULL plugin/hook adapters for DSH, Claude Code, Codex, and Pi, plus a host-agnostic SOFT skill with ADAPTIVE adapter guidance.
Cloud workspaces for DeepSeek Harness — pick Cloud (SSH) in the workspace picker and the agent's bash/read/write/edit/glob/grep tools transparently run on your Linux server over SSH. Zero remote install.
Zero-touch runtime telemetry for DeepSeek Harness: every session writes its own report card (task/model/tools/failures/duration) into a human-readable journal. 纯观察者运行事实遥测。
Auto-Approval: a middle permission tier for DeepSeek Harness between workspace-write and danger-full-access, auto-approving harmless commands and trusted-area targets
多文件夹 workspace：让 DSH（DeepSeek Harness）的 Agent 不只能读写主目录，还能同时读写你添加的其他文件夹。Multi-folder workspace for DeepSeek Harness: let the agent read and write several folders at once, not just the primary one.
Writes live provider model lists into DSH settings. Composer rings show 5h/7d plan windows or metered balance for the current session model. / 把线上模型列表写进设置，输入框圆环显示当前会话的 5 小时/7 天窗口或按量剩余余额。
DeepSeek Harness 自动审批插件：在 workspace-write 沙箱之上叠加确定性规则 + LLM 安全审批，自动模式不放宽沙箱、fail-closed。 Safe auto-approval for DeepSeek Harness — deterministic rules + LLM review on top of the workspace-write sandbox. Auto mode without ever granting full-access.
dsh plugin: rewrite your draft into a better prompt with the LLM and write the result back into the input box; optimization directions live in an editable skill, editable from the settings UI
Project-level change routing and reuse governance for AI coding agents. Helps Codex and Claude Code find the right capability boundary, avoid duplicate implementations, respect write constraints, and calibrate routing decisions for large full-stack repositories.
