Plugins

Vector + graph memory backend: observe/remember/search/recall tools against a local memoria server, fusing HNSW semantic recall, FTS5 keyword search, and knowledge-graph signals via RRF ranking, with automatic turn-end persistence and per-agent namespace isolation.

Microphone voice input for the composer: browser Web Speech API live transcription, dedupe/auto-continue, smart punctuation, language and auto-send settings.

Windows computer use for DSH built on a native C# engine: text-only DeepSeek models can read UI Automation controls and fill forms in verified batches without screenshots, stale control refs can be re-found by unique identity, CAPTCHAs and verification codes are refused and handed back to you through the DSH approval prompt, and Esc stops the turn at any time. A derived project based on Yu-tao-Li/dsh-computer-use-win; do not install it together with that upstream plugin, since both drive the same desktop.

A fork of dsh-livestream (MIT) - install it instead of that plugin, not alongside it. Lets DSH host a Bilibili livestream: chat and gift replies, queued viewer video requests on a skinnable stage, Doubao TTS voice, and automatic OBS/Bilibili streaming setup; needs OBS Studio 28+ and a Bilibili live room, and the voice stays silent until you add a Doubao TTS API key (Python + faster-whisper optional for transcribing subtitle-less videos).

Turns a repository and a software name into Chinese software copyright registration material - the 60-page source code excerpt (50 lines per page, auto-paginated), the software manual with screenshots, and the derived application-form fields (line count, language, completion date, function summary). Requires Python 3 with python-docx, and Google Chrome for screenshots.

Breeze TTS for DeepSeek Harness: read replies aloud in full or as a short summary, one-click Breeze CLI login with automatic key import, voice search and design, smart companionship, and lunch/dinner reminders.

Multi-model layered pipeline for DeepSeek Harness: a strong model plans, a fast one drafts, a third reviews, and a failed review is reworked automatically — endpoints, protocols, models and gates all configured from the DSH settings panel.

Three-position toggle (off / this turn / always-on) for an "unlimited mode" that injects a "do not conserve tokens" instruction into the system prompt; the "this turn" mode reverts automatically on the kernel agent/status idle event.

Runs a project's Claude Code opening questions and plan, hands the task to DSH, and offers post-turn Claude review choices. Windows only, and requires a locally installed Claude Code CLI (set claudeCliPath if it is not auto-detected). New sessions handed to DSH are switched to danger-full-access.

Roleplay companion memory for DeepSeek Harness: per-conversation editable persona documents (1-100 weight per entry, AI may refine them in small story-driven steps), per-speaker favorability and bond in multi-person chats, active memories with AI-scored vectors, a nine-emotion table updated every turn, an AI self-evolution ledger, recent 48-hour transcript search, companion-style conversation compression, a date-archived local knowledge base with auto-packing by turns or tokens, and a unified integrations page that lists curated plugins and downloads their GitHub releases.

Manages a Magpie model list and writes the checked models into a Magpie provider group: full sync from /v1/models, manual entries, per-model context, output limit, image input and reasoning efforts, sorting and drag reordering. On Windows x64/ARM64 it also downloads and upgrades Magpie from yetone/magpie-releases with SHA-256 verification and rollback, auto-upgrades once a release is 30 minutes old and all conversations are idle, starts Magpie in the tray with DSH, and toggles its lite mode and sleep prevention.

Picks each subagent model from the local catalog by task category, unless the caller already named one. Free and quota tiers are ranked together by per-model capability tags; a pricier model is used only when no cheap model is tagged as good at that kind of work. Only providers whose remaining balance is readable, or that are account-backed, enter the candidate pool; the rest stay listed but are not routed. A panel adds per-model manual priority above the automatic order plus allow/revoke controls.

Connect a self-hosted Magpie gateway with one provider route per upstream owner, so the stock model picker shows Magpie (Claude), Magpie (Codex) and other groups, plus automatic multi-line failover, read-only usage cards, per-model default effort, a per-session fast switch for priority-tier models, and optional Cloudflare Access.

Adds a browserless deepseek-web provider that drives chat.deepseek.com, with an account pool (rotation, cooldown, auto-failover), in-panel login (browser, or email + password with MFA) and token usage stats.

Automatically retries model failures (drops, timeouts, rate limits, 5xx) in-turn. Retry count, delays, backoff, fallback and more are configurable. Rate-limit backoff is capped and jittered, server retry hints are honored, and optional retry-until-success keeps going within a time ceiling. Every option is editable from the plugin config or the sidebar panel: retry count, initial and max delay, backoff factor and jitter, rate-limit delay window, retry-until-success ceiling, fallback trigger and switch cap, dead-route blacklist TTL, log file. When a route keeps failing it can switch to a pinned, allowlist-filtered, or catalog-picked alternate model route, with a 30-minute dead-route blacklist (auth failure, subscription expired, billing) and a stale-assembly guard that keeps retried requests on the switched route. Ships a right-sidebar panel with live retry state, route labels and incident counters, Chinese and English locale, and memory-only logging by default.

Composer control for the automatic context-compaction threshold. A 20%-80% slider rewrites the running BasicCompactionEngine config per step, shows current usage, context window and the effective trigger line, and compacts immediately when the threshold drops below current usage and the agent is idle.

DSH sidebar and panel for courses, tasks, plans, connections, course materials, automation and settings.

Whole-disk file search tool backed by the local Everything (voidtools) index: millisecond path queries for the agent over a loopback-only HTTP fast channel with es.exe fallback, plus verified auto-install of the latest stable Everything + ES CLI backend (signature-checked, proxy-aware). Windows only.

Saves a whole DSH session as one .dsproj project file and opens it back. The session header gains an open button, a save button, a save-as button and an autosave switch, and the blank-session page gains its own open button. The first save asks where to put the file and later saves overwrite it; opening a .dsproj re-imports the session, switches to it, and can write the bundled workspace memory back on request.

Autonomous AI assistant and multi-agent collaboration for DSH: discuss goals, plan work in the background, coordinate multiple Workers to execute tasks and review results, pause and resume work, recover after interruptions, keep task history and deliverables, and build working memory and skills; offers a standard local mode and optional experimental private-Git collaboration.

Douyin (抖音) automation for DeepSeek Harness: watch private messages and wake the agent event-driven, plus 10 tools (read DM, send DM, comment, @ mention, reply, delete comment, read notice, share video card, pick feed candidates, cookie backup/restore) —

Project autopilot for DSH 0.2.1-rc.2 and 0.2.1-rc.3, with scoped authorization, independent verification, editable memories and companion chat, including automatic installation of the compatible desktop pet fork. The bundled pet is a fork of takboo/dsh-coopanion (AGPL-3.0-or-later); do not install the upstream pet alongside it.

Session-level autonomy gears for DSH: a rail on the edge of the conversation area switches P/R/NT/NC/NE/M-/M+/D/A and injects the active gear's behavioral prompt into every request; the model's set_gear tool can only downshift, and handbrake H blocks all tool calls. The set_gear tool can downshift on its own; upshifts need user approval.

Tab autocomplete for the DeepSeek Harness desktop and web composer: next-message ghost text, inline draft completion, and Alt+Up/Alt+Down candidate cycling.