Plugins

All-in-one DSH extension center in Settings: manage skills, MCP servers and plugins, a Plugin Market (curated + GitHub search, npm one-click install), and an Add-ons block to install, disable, uninstall and update companion features together.

Cost tracking priced from an online catalogue (models.dev + OpenRouter, 8000+ models, each call priced at its own instant): a per-turn cost line, a Cost tab attributing spend to tool output / model output / system prompt / commands, budget, monthly forecast, ~166 live-rate currencies, and a bill_stats agent tool.

Auto-migrate Codex desktop-pet skins into DSH: an animated corner pet that mirrors agent state (thinking, tool use, awaiting approval, failed, done), with per-session frosted-glass dialogs and a full settings panel.

Runs open-mcp-apps inside DSH: every MCP app gets its own sidebar container with a separate workspace, session, and App mode, an agent status strip under the app, and inline app rendering in ordinary chats.

为 better-sidebar 提供 Office 三件套预览（.docx/.xlsx/.pptx），独立 bundle 瘦身主体 | Provides Office-suite preview (.docx/.xlsx/.pptx) for better-sidebar as a separate bundle to slim the core

Model-based permission approval: an approval-request answerer backed by a separate reviewer model.

Bring the deepseek.com web/app chat experience to DSH: edit your prompt and regenerate answers in place, with a per-message <i/N> version pager (tree model, stable across conversations).

Web appearance suite with a 7-theme gallery (each a full 122-token --dsw-* mapping) plus one-click switching persisted in settings, Material You wallpaper color extraction that generates a whole dynamic palette, and a global corner-radius control. Themes ride on the official look system as override layers instead of fighting it, so the official light/dark/auto row stays valid. Frozen archive: keep the pinned version, as the author no longer adapts it to newer DSH releases.

Registers the minas-webdav MCP server as one stdio row, giving the agent 14 tools for a Xiaomi Smart Storage (MINAS) NAS without mounting it: list, stat, tree, find, du, df, cat, upload, download, mkdir, move, copy, delete and a capability probe. The Python package behind it is installed separately with pipx.

Agent presets whose leader does no work itself: every non-management tool call is denied at the harness boundary so the task has to go to a subagent, after a one-time lookup of the available subagent models and a per-task model choice; a second preset deliberately picks the model least suited to the task.

Web appearance suite with a 7-theme gallery (each a full 122-token --dsw-* mapping) plus one-click switching persisted in settings, Material You wallpaper color extraction that generates a whole dynamic palette, and a global corner-radius control. Themes ride on the official look system as override layers instead of fighting it, so the official light/dark/auto row stays valid.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models. Within a turn it supports skeletonized information compression with a low original-text retrieval rate; at the tail of a turn it supports a turn-tail intent summary implemented purely by content substitution; context content stays safe, and the whole process runs inside the official JSONL compression.

Replaces the stats line under the message input in the DSH desktop app with a configurable status bar of ten optional fields, with live cost estimates at DeepSeek's official peak and off-peak rates, custom per-model prices, a CNY or USD switch and account balance.

Extends the working range of DSH into WSL. Files in the distribution can be read, written, edited and searched by DSH directly. Depending on the task, the AI chooses for itself: run a one-off command, or open its own terminal and run several commands there to finish the job. Typing `dsh` in a WSL terminal calls DSH (type `dsh help` there for details), fusing WSL and DSH deeply. Every terminal you open can collaborate visually with the AI.

Rotates configured LLM API keys after quota, rate-limit, or authentication failures.

AI interview practice for DeepSeek Harness Web: generates questions across four modes (concept review, mock interview, scenario, LeetCode Hot 100), scores answers, and provides explanations.

Dead-code detection for AI-generated codebases: graphlint builds a dependency graph, finds code unreachable from any entry point to enable codebase cleanup and functional‑effectiveness understanding, and exposes graphlint_query / graphlint_build / graphlint_config tools plus a graphlint skill to DSH agents.

Configurable DSH usage statistics: a sidebar text/card widget plus a detail dialog with per-provider and per-model breakdowns, cache-hit rates, token trends, and latency details (TTFT and duration).

Office workspace & session dashboard: a floating 6-column sprite panel visualizing workspaces, sessions, token usage and subagents, plus Agent Mail and an office log tab. Feishu/Lark views (message feed, meeting schedules, transcripts) additionally require the `@larksuite/cli` tool installed and authenticated separately.

Windows XP Luna desktop for the DSH Web UI: a floating window manager with taskbar and desktop icons, plus the era-accurate Luna skin.

Fifteen curated theme families plus custom CSS-only themes with import, preview, apply, delete, and restore-default flows that follow the native Light, Dark, and Follow system modes.

开窍 Aha Tutor: a personal AI tutor for DeepSeek Harness that builds a course around you and teaches it on a narrated handwriting whiteboard

Brings the WorkBuddy, Trae and Qoder desktop logins on this machine into DeepSeek Harness as OpenAI-compatible local endpoints, with an account pool, automatic account switching when WorkBuddy is rate limited, and a settings panel.

Image generation through a local ComfyUI server: one generate_image tool covering text2img, img2img, inpaint and background removal.