Plugins

Turns corrections into permanent rules: explicit remember: triggers and auto-detected corrections pass a confirm bar (save globally / per project / edit / skip) and are written into a marked AGENTS.md section. New rules take effect in the current session and load natively in future ones; writes are deduplicated and version-guarded, and secret-looking content is refused.

Adds a Timestamp toggle to the composer; while it is on, each message you submit is followed by a hidden context message carrying the current local time, so the model knows when you sent it. The switch is remembered per conversation.

Imports wallpapers newly downloaded by the Hao Wallpaper app into the DSH wallpaper engine on startup, incrementally and idempotently. Requires the Hao Wallpaper app, Python 3.8+ and the dsh-plugin-wallpaper-engine plugin (elysia395/dsh-wallpaper-engine) installed first. Supports macOS and Windows.

Enlarges the DSH web settings dialog (default 1180x920, 20px steps, remembered per browser) and adds an Advanced Settings page that resizes it and hosts Host operations: restart dsh with a 3-second delay and a busy confirmation, restart once all sessions have been idle for 10s (cancellable, repeat clicks idempotent), last start time with uptime, and a dsh version/update check that compares npm dist-tags across selectable channels without upgrading. The same bundle replaces the sidebar brand area with a machine name, adds a read-only provider, model and thinking-level cell beside the composer stats pills, and a switch that registers or unregisters the three dsh_restart_* agent tools at runtime. Zero dependencies, no build step; all routes sit behind the connection trust fence.

Adds a full-screen dialog for writing long messages: Markdown on the left rendered live by marked on the right, with a draggable splitter that remembers its ratio. Confirming writes the text back to the composer draft, keeping command tokens and @-reference chips as a fixed prefix; Esc asks whether to discard or write back before closing.

ReMe-shaped long-term memory for DSH: Markdown workspace, BM25 recall, turn capture, scheduled consolidation

Moves and resizes the DeepSeek Harness settings panel; its position and size are remembered per browser.

Whale-girl desktop pet for the DSH Web UI, drawn with hand-written SVG + CSS (no image or audio assets, zero runtime dependencies): swims while a turn runs, blows labeled bubbles on tool calls, taps the glass with an amber pulse when an approval is waiting, jumps with a turn duration/tool-count report on success and droops on failure, naps after dark or when idle. Click to poke, double-click to pet (affinity hearts), drag with remembered position, right-click menu (feed / mute / reset / hide), hover card with level, title, XP bar and per-turn stats. A `whale_say` tool lets the agent make her speak or act as a lightweight visible notification. Read-only over sessions and workspace; only ~/.dsh/whale-girl/state.json is written.

Draggable floating TES-gyro widget for the DSH Web UI: click to spin it with linear friction decay, tune the spin speed from a slider, and play the bundled track with its LRC lyrics scrolling in sync. Spin speed and position are remembered.

Test work framework for DSH that scaffolds a product, module and version directory tree, registers seven test_framework_* tools and four runtime skills, and adds a prompt-level gate that reports which requirement-document splits, versions, related documents and environment credentials are still unrecorded before testing starts. Requires the playwright MCP server (no fallback).

Project Development Mode preset plus an in-chat development board for DeepSeek Harness: the lead conversation clarifies requirements and delegates to an Agent Team, and the floating board tracks each user change request through dispatch, execution, testing and acceptance with 7 board tools.

Skill-creation console for the Web UI: fetches the official creator skills, hands the requirements to the Huawei Cloud, generic, or REST API creator, then scores the generated skill against a 10-point rubric and returns the itemised deductions to the creator until it reaches 8.5/10.

Evidence-first reuse skill for DeepSeek Harness: before implementing, porting, choosing dependencies, or troubleshooting a tool, it searches the local project and machine for existing solutions in layers, judges reuse on checkable evidence, refuses reuse when essential requirements or risks are unmet, and reports what could not be verified. Ships as a DSH bundle that registers the skill on ctx.skills.

OCR for scanned, image-only PDFs into Markdown, built around the fact that a book takes about an hour: every page is checkpointed so an interrupted run resumes, a progress log reports live status, and a quality report names empty pages and low-confidence lines. Requires Python 3.10+ and pip install -r requirements; the first run downloads the OCR model.

Search public-domain libraries (Project Gutenberg, Wikisource, Internet Archive, Standard Ebooks) by title, author, character or half-remembered plot, or import EPUB / PDF / TXT / DOCX / Markdown, and listen in the sidebar with Microsoft Edge neural voices, offline Kokoro or the device voices; keeps your position, takes voice notes anchored to what you just heard, and discusses the book with your DSH model.

Registers two tools for DeepSeek Harness. `doc_ocr` turns scanned PDFs and images into structured text (per-page text, financial tables with a balance-sheet tie-out check, contract-clause hits) through a local RapidOCR engine or a Zhipu GLM-OCR cloud engine, and `doc_preprocess` turns a directory of raw files into an Obsidian-ready knowledge base through a seven-stage pipeline with backend routing and sha256-based incremental skipping. Numbers that fail the tie-out are reported, not invented. The local engine needs Python 3 with rapidocr-onnxruntime, pypdfium2 and Pillow (pip install -r requirements.txt); the GLM engine needs ZHIPU_API_KEY.

Structured image recognition: local sharp measurements plus a cloud vision model return description, OCR with coordinates, object locations and fine-grained attributes in one call; a fingerprint cache makes repeated images return in milliseconds without spending model quota.

One switch that takes an MCP server's tool schemas and its instructions block out of the model context for every agent, live and without a restart, built for the 207 tool schemas REA registers whether or not a task is reverse engineering. A pill in the composer dock, /rea on|off|status|toggle, and GET/POST /rea-toggle/state; the position is remembered across restarts and subagents are covered before their first step.

Adds a Background button above Settings in the sidebar foot that fills the window with one of six preset wallpapers, the theme default, or a custom image URL, remembered across reloads.

Ctrl+wheel page zoom (60–180%, remembered) for the whole DSH UI, plus a bottom-right Settings entry that opens the host settings panel, with Ctrl+0 to reset.

Freezes the DSH chat scroll position: new replies no longer drag the viewport to the bottom — it stays where you left it. Only after you send a message does it follow to the bottom so you can see your own input, then it locks again once the answer starts streaming. Each session remembers its own position (persisted across reloads), a settings toggle turns it off, and it never intercepts your own scrolling.

Global back/forward navigation for the DeepSeek Harness Web UI: remembers every session and whole-page panel you visit, shows a configurable 1-7 entry history strip you can click, and supports browser-style shortcuts plus mouse side buttons.

A piano keyboard in the DeepSeek Harness UI: click a key to play it, or hold and drag sideways for a glissando across the keys. Notes are synthesized live with Web Audio — no samples, no network, no third-party dependencies — and the volume and starting octave are remembered between sessions.

Daily brief for DeepSeek Harness. Every day at a fixed hour it ranks the plugin market by real install counts and stars, cross-checks it against your own agent memory and working conventions, and proposes two kinds of item: concrete deltas on the plugins people actually use (original does A, the proposal does B) and directions nobody has built yet. The popularity numbers come from the catalog the market itself publishes and are quoted into the prompt rather than guessed; an entry the market has not measured is shown as unmeasured, never as zero. Approving an item opens a new session named after it and hands that session a fixed delivery contract — implement independently, never fork, test green before publishing. Rejected sources are remembered so they stop coming back.