Plugins

Parent-agent dispatch with review: split a task into work orders, hand each to a contractor you picked (its own provider, model, and reasoning effort), then pass or fail every delivery — failed ones are re-dispatched with your feedback.

Adds two audit tools and four skills for Chinese medium-to-long texts (fiction among them) and document review — seven-dimension AI-flavor scoring with line-numbered blocking findings, plus file-backed continuity state for long manuscripts. Do not install with the older dsh-deai-audit from the same author, which registers the same tool names.

Chat-driven image-deck workbench for DSH: intent to storyboard script, per-page prompts and AI image generation with page-level revisions, nine design-style presets, and PDF/PPTX export (one full-page image per slide).

Local GPT-SoVITS text-to-speech for DSH: reads replies aloud, switches voices from the right sidebar, and trains a new voice from a dropped audio sample. First run downloads an 11.4 GB engine and needs an NVIDIA CUDA GPU.

Give DSH AI timestamped video frames, audio and lyric analysis for Bilibili, YouTube and local files, with encrypted browser-session import, continued viewing, saved site resolvers and native cache management. Requires FFmpeg and ffprobe on PATH or configured in the plugin settings.

Mobile-viewport preview tab for the right sidebar, with nine preset sizes covering phones, tablets and a laptop plus a custom width/height, one-click URL copy, and the page URL inserted into the composer through an @ button or the /url command.

Agent Settings: manage reusable sub-agent profiles in the sidebar, each with a when-to-use description and its own provider/model route, and let the agent start the matching one as a subagent.

Manage $DSH_HOME/.env from Settings and sync Host plus Windows/macOS user environment.

Adds a balance-checking tool, injects a per-turn DeepSeek account balance snapshot with a low/critical power-saving and stop policy, and provides a settings page for thresholds and FX-linked currency conversion. Requires a signed-in DeepSeek account for balance reads; API-key-only profiles return an unreadable result.

Local knowledge base for DeepSeek Harness: distills workspace and chat-attached PDF, Word, Excel, Markdown, TXT, CSV and HTML documents into Markdown notes, with a VS Code-style viewer, per-workspace access grants and agent tools to search, read and add notes.

Learns a user's coding conventions from instructions and corrections, stores them as confidence-scored rules in editable Markdown, and injects them as reference data into every session. Rules that decay out of range or contradict a newer rule stay visible for review instead of being applied.

Persona plugin: a slowly moving 0-100 emotion meter, its own worldview, a per-turn diary, a relationship network, proactive messages, and an inner-voice line beside the cost row.

Adds a connection-test panel under Settings -> Models: reads the deployment's model catalog and probes each configured model with one minimal chat round-trip (a short "ping" with a small output cap) through the Host LLM service, verifying connectivity with a real request. Each row reports success/failure with the harness failure code (NO_ADAPTER, MISSING_CREDENTIAL, AUTH, RATE_LIMIT, TIMEOUT, ...) plus first-token/total latency; single-model and concurrent test-all (3) are supported. Credentials are resolved inside the Host process and never reach the browser; probes are independent per model behind a same-origin-fenced route with a body-size cap. Pure JS, no build step, MIT.

Adds Meta's Muse Code subscription as a DSH model provider: the bundle declares the shipped Meta route under Settings > Models, syncs the account chat catalog (Muse Wonder included) into the model picker, and adds a Sign in card plus the /muse-login command that run DSH's built-in Meta device-code flow, keep the minted key renewed, and show a red or green credential dot. Needs an active Muse Code subscription and no Meta CLI.

Search-first model selector with favorites, recent models, collapsible provider groups and keyboard navigation.

Registers the models of a local workbuddy2api-hub gateway into DSH as one OpenAI-compatible provider per gateway API key, each key keeping its own model selection, and shows the main key's credit balance plus today's and this week's tokens, credits and API-equivalent cost in a sidebar badge. Requires workbuddy2api-hub running locally.

Acid-graphics theme for the DeepSeek Harness Web UI: six complete palettes (acid, neon orange, electric cyan, violet, rose, warm orange-light) registered as switchable themes, plus an optional WebGL wireframe crystal backdrop driven by pointer and scroll, local wallpapers with a dimming mask and background lettering, a crystal startup animation with optional local video, and a Settings section that holds every control.

A deep-sea dive startup splash for DeepSeek Harness: a depth gauge with a submersible probe, a character reveal video that then loops, and a seafloor with coral, bubbles and slow drifting particles. The depth reached decides which sea creatures appear.

A liquid-glass skin built on your current desktop wallpaper: the wallpaper is blurred and veiled into the window canvas, panels stay translucent, and every surface that carries text (composer, code blocks, preview pane, menus) gets its own frost level so nothing is washed out. Apple-style liquid-glass controls. On Windows the canvas reads the current wallpaper from APPDATA; other platforms fall back to a painted gradient.

Color presets and a custom paper-color picker for the built-in DSH sidebar PDF, converted Office, and spreadsheet previews, with dark-mode brightness control and locally saved preferences; changes only the display, not the files.

Per-turn and per-step cost ledger for DeepSeek Harness: a Turns tab beside Chat and Trajectory with tokens, estimated cost, cache savings, and an inspector for tool arguments, raw results, and source. Client-only; prices are estimated from recorded model usage.

DeepSeek Harness activity as a GitHub-style contribution graph: an account-menu row opens a centred dialog with your avatar and name, a year of daily activity in five shades, a year picker and a tooltip on every day.

Forensic analysis of past model-routing decisions. Replays each session, classifies every turn's task complexity from its request shape (deterministic, no LLM), and reports two kinds of waste: an expensive model on cheap work (the difference is measurable), and a cheap model on expensive work that failed on its own merits — where every retry re-sends the whole context, so the downgrade cost more than the upgrade. Distinguishes provider-capacity retries (RATE_LIMIT/TIMEOUT/TRANSPORT) from model failures so it does not cry wolf. Derives routing rules from the evidence, priced against a built-in per-million table.

Lightweight sidebar UI for DeepSeek Harness: five fold switches, three density presets and three width presets, served from one footer row plus a settings section — no dependencies and no background service.