95 results
DSH plugin: fixes compaction failure on local qwen3.8-27b gateways — xhigh thinking burns the entire output token budget, so thinking is off for compaction-only, with the model's non-thinking sampling parameters
Force DeepSeek Harness agents to think, reason, and reply in Chinese — a one-section dsh bundle
DSH plugin: detect long model thinking chains, segment them, and show per-segment summaries live in the chat UI, with a settings page.
Matrix rain for expanded thinking in DeepSeek Harness Web
DSH 自定义模型思考等级管理、渠道重试策略与偏好记忆插件：原生设计语言、全档位思考配置、渠道级重试控制与无感隔离式跨会话偏好记忆。
第三方模型思考努力度配置（DeepSeek Harness 插件）：在设置页为模型声明 reasoningEfforts 档位、路由默认档位与 Anthropic token 预算，写入 llm-pi-ai 命名空间；档位调节复用原生模型框 Effort 面板。
A lightweight plugin that adds model thinking token statistics to the bottom Dock and the end of each conversation.
OpenAI-completions-compatible adapter for custom gateways (vLLM / LM Studio / self-hosted proxies): always role:"system", enable_thinking / chat_template_kwargs / reasoning_effort driven by the llm-pi-ai model config, and a </think> split on the receive side — no pi-ai behavior guessing.
Live reasoning-token badge — Think(5.2K) — on every Think disclosure row of the DSH web UI, ticking while streaming · 在 web 界面 Think 行标题后实时显示思考内容 token 数的 DeepSeek Harness 插件
DeepSeek 中转站（OpenAI 兼容网关）思考强度适配插件：让 UI 出现 Off/Low/High/Max 推理等级选项，并按中转站格式发送 reasoning_effort / thinking
Per-turn thinking-token statistics for DeepSeek Harness, folded once by a durable session projection and read by the composer dock.
Lightweight personalisation for the DeepSeek Harness running-turn status line: rotating bilingual quips, six configurable loading icons and one-click theme-colour matching. A drop-in DSH web client plugin (bundle) with no dependencies and no build step.
让 DeepSeek Harness 的 Agent 用简体中文思考和回复（全局插件，对所有 preset 生效）
DSH 中文思考开关:设置页开关,引导模型用中文进行内部思考
Extra prompt slots (${provider} ${thinking} ${time} ...) for the official turtle-ui TUI of DeepSeek Harness
An Eason-themed floating desktop pet for the DeepSeek Harness Web UI: 18 Eason-only transparent animations (concert, coding, daily life), session-linked behavior mirroring the agent (thinking→coding, streaming→singing, tools→progress bar, waiting→peeking,
DSH bundle: plays distinct chimes in the Web UI on message received, thinking start, tool call, and turn end.
LongCat (LongCat-2.0) model provider for DeepSeek Harness — 1M context, thinking mode, tool calling
DeepSeek Harness tools for external opportunity discovery, unmet-need research, pain-point signals, opportunity mapping and validation experiments.
Model selection as a three-level cascading menu (Provider -> Series -> Full model) plus thinking-intensity (reasoning effort) selection for the composer model seat.
Google Antigravity (agy CLI) models for DeepSeek Harness — stream Gemini/Claude/GPT-OSS subscriptions into DSH with thinking, tool activity, token usage and in-GUI Google OAuth login.
DSH Web 思考强度就地替换插件：把模型座位「推理等级」面板里的原生单选行替换成 Claude Effort 卡片风格的美丽滑块（Off / Low / High / Max，含 WebGL 火焰），选择仍走官方会话链路。
DeepSeek Harness web GUI: collapse the agent trajectory (tool calls, thinking, workflow steps) while always keeping the agent's final output. A setting controls whether the trajectory is default-collapsed after the final output completes (default on).
DSH client plugin: Think token meter + fold cards grouping thinks & tool calls between text dividers.