플러그인

Every few turns asks the model whether anything is worth keeping, then sorts the answer into memory, skill or hook.

Brings Google Antigravity (agy CLI) models into DSH with streaming, thinking and native tool cards, a manual multi-account pool with quota and proxy controls, and an optional experimental direct transport. Community fork of amlyczz/dsh-agy-link. Google's Antigravity FAQ says personal plans do not support wrapping agy in third-party coding agents, so using it may put your Google account at risk of restriction. Needs Node.js 24 or newer and Google's agy CLI on PATH; the experimental direct mode is off by default.

Adds a Plan/Build switch to the composer chip and to the Tab key, keeping a separate model and reasoning effort per mode and restoring them on every toggle, /plan, and plan approval.

Takes Enter away from sending in the DSH composer: Enter inserts a line break, every Ctrl/Cmd/Alt+Enter variant is swallowed, and messages leave only from the send button. A gray hint in the input box says so before the first keystroke and stays beside the button while typing.

Fully background browser automation and browser use for DSH: an offscreen Electron window per conversation with real CDP input and screenshots, so headless-style browsing never steals focus or pops windows. Runs on Electron and keeps cookies in a persistent user-data directory, so a signed-in page stays signed in for later conversations.

Externalizes oversized assistant and tool messages in long sessions into project files while keeping a recent window, and archives reasoning from empty turns as a visible message so long tasks do not stall on context size.

Colour-theme and wallpaper plugin: 62 colour themes (37 built-in, derived from OpenCode palettes, plus 25 IP presets), each shipping its own Markdown and syntax-highlight palette, a layer that pushes sub-4.5:1 semantic colours back to readable contrast, and local background images with fill or tile, blur, scrim and rotation. Replaces the earlier cnhenaihui/dsh-ios-theme.

Adds a notepad to the DSH Web UI that floats as an edge-snapping, resizable card or docks into the real bottom layout row, switchable at runtime with shared text and images; notes auto-save (text in localStorage, images in IndexedDB) and follow the active theme.

Lets the agent drive the Chrome tab you @-mention — 13 tools for tabs, snapshots, clicking, typing, keys, scrolling, dragging and screenshots, with input sent as real CDP events so pages see isTrusted true and a pointer drawn inside the page. Needs the bundled Manifest V3 extension loaded by hand in Chrome (developer mode; it requests the debugger and all-URLs permissions), talks to it over a WebSocket on 127.0.0.1 ports 30795-30805, and saves screenshots under ~/.dsh/chrome-tab-shots; the macOS native input channel also needs Accessibility permission and the Xcode command-line tools.

Checkpoint-and-resume for interrupted runs: when a conversation reaches a key step — two consecutive successful file writes/edits, or one successful test command — it appends a lightweight record (timestamp, step type, tool, ≤200-char argument summary, result status) to a local JSONL file, keeping the newest 50 and pruning after 24 hours. After DSH restarts, the session_recovery tool lists unfinished sessions and resuming one creates a new session whose system prompt carries only the last three checkpoints (≤500 characters) instead of the full transcript.

Puts a domain bar under the session header — a domain groups skills and tools and toggles as one block, so one click loads or unloads every member. Hovering a chip lists its members (skills, then tools) in a height-capped panel that scrolls. The right end of the bar is a compaction-threshold counter (− 5 +, 100k tokens per step) that rewrites the live compaction engine policies, so the compaction trigger line changes on the next turn without restarting dsh; a degraded fallback turns the number amber and is reported in the snapshot. Settings maintains domains and members with search, skill/tool tabs, all/selected/unselected filters and a copyable AI grouping prompt. Tool removal uses tools.restrict plus a guard; hiding a skill re-registers it with modelInvocable false, so it leaves the model catalog while /name still works. Sessions of another preset are left untouched.

Arknights-style procedural opening film and silent looping chat backdrop for the DeepSeek Harness Web UI: Canvas-drawn visuals, synthesized WebAudio sound, click-anywhere skip, cross-dissolve into the interface, live-adjustable opacity; zero dependencies and no media files. By default the film (about 12 seconds, with sound) plays on every page load; a setting turns it off.

Turn the seam between the sidebar panel list and the workspace list into a divider you can drag, snapping one row at a time, with the panel list scrolling when squeezed.

Visualizes Finesse 3 KatScript in the conversation and adds a drag-and-drop optical bench designer that generates the Finesse model. It scans every user and assistant message for KatScript or a Finesse Python cell and renders them in the background with the first Python interpreter that can import finesse (Finesse 3, NumPy and Matplotlib required; interpreter overridable with FINESSE_VIZ_PYTHON), keeping the code and figures under ~/.dsh/finesse-vis.

Composer for the DeepSeek Harness Web GUI: bracket and quote pairing, wrap-on-selection, closer skipping, empty-pair backspace, and a completion popup over draft words, language tables, workspace files and model ids. Full-width brackets typed through a Ch

Declare local executables, scripts, bash and PowerShell commands as Agent tools with exact argv mapping, hot reload and a settings page.

A desktop pet for DeepSeek Harness: an always-on-top character and a roster panel that show what the watched conversations are doing, with seven animations, Workspace grouping, subagents drawn under their parent, and one click to open a session. Ships Windows, macOS (Intel and Apple silicon) and Linux window binaries in the package, so installing runs no install scripts and needs no build step.

Cat desktop pet for DeepSeek Harness (dsh-coke): click it for one of 7 catchphrases with a springy bounce, right-click for chat (answers via your current model, optional sticker), singing with a microphone, a random 'form' card, green-screen outfit swapping, or shooting it; balance query at its feet (needs a DeepSeek API key); auto-starts with DSH. If no Electron runtime is found it first downloads about 100 MB into ~/.dsh/electron (in-page mode works meanwhile).

Replaces identity numbers, phone numbers, bank cards, emails, addresses, customer names and secrets with numbered placeholders before a model request leaves the machine, restores the response, and keeps an encrypted local mapping plus a hash-chained audit log exportable as CSV.

Toolbox panel for the DeepSeek Harness Web GUI: a right-side drawer (full screen on phones) with connection information (Wi-Fi details and gateway ping need a Windows host), one-click diagnostics, a speed test with server-side timing, a quality monitor, history and auto patrol; the host registers the /api/toolbox routes.

AI-assisted research workbench with editable step workflows, on-demand skill binding, a live execution graph and in-project document review (needs a local Python with python-docx, python-pptx, openpyxl and PyMuPDF), while keeping existing projects on their original configurations. Requires DSH 0.2.2 or later.

DeepSeek Harness port of the Impeccable frontend-design skill: 24 design playbooks covering shaping, critique, audit, polish, layout, typography, motion, accessibility and design-system documentation, driven by the reference files shipped in the skill plus PRODUCT.md / DESIGN.md in the project instead of a downloaded engine binary. Registers the impeccable skill; reads project files and writes design artifacts under .impeccable/. The binary-only features (live browser bridge, pixel measurement, font matching) fall back to static variants, screenshots and explicitly declared manual estimates. Ported from pbakaus/impeccable (Apache-2.0).

A document generation plugin for DeepSeek Harness: JSON is the core output, structurally managing document content; unifies parameter definition and inheritance, material extraction, validation checking, and scoring across document generation, keeping output on-target, fast, and token-efficient; ships a built-in COSMIC scenario pack as a worked example plus a built-in guide for authoring new scenario packs.

Grabs a live frame from the local webcam and hands it to the agent as an image, so the model can actually look through the camera without opening any app or keeping a window in front. A dim room no longer comes back black — brightness, gamma, contrast, sharpen and an auto-level pass run inside the capture itself. ffmpeg is fetched automatically on first use, and the same plugin lists the available capture devices.