插件

调整 Web UI 左侧会话列表栏宽度：调低 264px 拖动下限，可选调整拖动上限与展开默认宽度，启动时自动修补 ui-layout bundle。

回形针助手回来了，这次它真的会干活：会对真实 agent 状态做出反应的办公助手宠物，任务完成时跳起来庆祝，回合失败时弹出经典的 "illegal operation" 对话框。

Web UI 局域网密码门禁：同网手机/平板输共享密钥登录，会话与电脑实时同步，内置 randomUUID polyfill。

Vision for text-only LLMs: image description / OCR / VQA via free Gemini and GLM vision APIs.

Session cost estimate in the DSH Web composer dock (tokenUsage × configurable price table, one-click official-price refresh).

Agent FinOps / token analytics for DeepSeek Harness: usage collection, time-aware cost engine, cache analysis, and session drill-down

Real-time code feedback for DeepSeek Harness — LSP, linters, formatters, type-checking, and structural analysis.

DeepSeek Harness IM bridge plugin: route Telegram & WeChat messages into per-peer agent sessions and reply back

DeepSeek Harness plugin: real-time Feishu/Lark notifications when a DSH session pauses for user interaction (permission asks, questions, fatal errors).

pi-style system prompt for DeepSeek Harness (dsh-pi). A Cordis plugin bundle that replaces the system prompt with pi's minimal-core, tool-listing, guideline style.

Web UI plugin for DeepSeek Harness: import skills from local Markdown files or URLs into the skill roots the harness discovers automatically

User Message Minimap: a navigation rail for long DeepSeek Harness conversations

Conversation-style UI/UX visual review plugin for DeepSeek Harness: vision_review / vision_ask tools backed by any OpenAI-compatible multimodal endpoint (e.g. agnes-2.5-flash)

Generate images and videos in DeepSeek Harness through the NanoBananaPro API

Generate images and Seedance videos in DeepSeek Harness through the Seedance 2 AI API

Runtime hot-swap for DeepSeek Harness: hot enable/disable/restart Cordis plugins and auto hot-mount bundles from the Web GUI — no dsh restart.

套餐用量角标：在 Web 对话框右下角显示 5小时/周限/月限 三个用量窗口（支持 OpenCode Go、GLM Coding Plan 双渠道：Z.AI 国际版 / 智谱，以及 Kimi Code 会员套餐额度，月度会员额度可经 kimi-auth Cookie 增强），并在「设置 → 插件」提供各渠道独立的启用开关与 API Key 配置。

DeepSeek Harness plugin for importing Codex App Server conversation history

多引擎反向识图聚合：Google Lens、百度、Yandex、TinEye、SauceNAO、IQDB、Ascii2d。

文件型工作记忆：memorize/recall 把关键前提逐字保存在会话笔记文件，无损挺过上下文压缩。

动作-状态时序记忆：记录类型化的状态与动作，做趋势、异常与因果关联分析。