Plugins

An agent preset for evidence-checked research: every step advances only through system-verified observations, L3+ claims go to an independent evaluator, and the session folds into a replayable ledger that grows into a browsable domain ontology. Requires DSH >=0.1.7-alpha.1.

Desktop pet (QQ-pet style): floats in the corner, draggable, feedable, playable.

The memory that dreams — cross-session memory for DeepSeek Harness: recalls you and your projects across sessions, consolidates memories while idle (autoDream), freezes conflicting memories for your review, and keeps a replayable audit trail. Memory stays on your machine as SQLite with human-editable Markdown mirrors; four-way recall (keyword, vector, BM25, entity graph) and an interactive memory panel ship with it, plus 10 model tools, an MCP server, an external API and a CLI. Summarization and consolidation run on the host model you already configured; with a local embedder, semantic search runs fully offline.

The memory that dreams — cross-session memory for DeepSeek Harness: recalls you across sessions, auto-consolidates in its sleep (autoDream), freezes conflicting memories for your review, and keeps a replayable audit trail. Offline-first local semantic search over SQLite with human-editable Markdown mirrors, an entity graph, an interactive memory panel, plus an external API and CLI.

The memory that dreams — cross-session memory for DeepSeek Harness: recalls you across sessions, auto-consolidates in its sleep (autoDream), freezes conflicting memories for your review, and keeps a replayable audit trail. Offline-first local semantic search over SQLite with human-editable Markdown mirrors, an entity graph, an interactive memory panel, plus an external API and CLI.

Drive OriginLab Origin scientific plotting from AI chat via MCP: 62 tools. Modal-dialog watchdog, LabTalk safety gate, release/reconnect, fit with initial/fixed/weighted params, fine-grained plot and sheet edits, matrix tools, FigureSpec declarative YAML (diff-able and replayable), MCP Resources read-only snapshot, matplotlib-figure bridge, PowerPoint assembly with panel labels, Graph Gallery template search, EPS export, file-access whitelist, visual regression baseline, and editable OPJU delivery with deterministic readback verification.

Auto thinking-effort routing for dsh: a locally fine-tuned 7-question classifier (ModernBERT, 842MB, <1s CPU inference) picks low/high/max for no-intent messages; in-session retries escalate the tier (low to high to max) under exclusion constraints; the judge service fails safe to low if unreachable. A tier chip in the input bar shows the live tier and the last 20 judgments. All inference is local - task text never leaves the machine.

Presents DSH Web as a playable, full-screen first-person Minecraft-style voxel interface—not a color-only skin: users walk through a persistent Three.js block world and access native workspaces, sessions, conversations, tools, model selection, permissions, context information, and the composer through in-world facilities and a workbench.

Zero-dependency, event-driven voice announcement plugin: no extra model, no token cost. Speaks with the system's built-in natural voice, supporting both Windows and macOS; final-reply announcements, approval & question alerts, optional event announcements (turn end, command done, goal change, tool errors, todo updates), replayable final replies, and a bilingual visual settings page.

Galgame generator: give it a script document plus character/background/BGM/animation assets in the workspace (img_human/, img_bg/, audio/, img_cg/) and it builds a playable visual-novel web page with configurable save slots (`[存档 N]`, default 9), speaker-following sprites with multi-pose variant switching (`[表情]`), opening/ending/CG animations (`[op]`/`[ed]`/`[cg]`), BGM play/stop/volume control, and if-line branching.

Desktop pet drawn from the official whale outline, for both the desktop app and the web UI: SVG animations and facial expressions follow agent state (thinking, working, done, error, waiting for confirmation), and it can follow other running sessions with a count badge; poke it, drag it, shake it dizzy, squash it against a screen edge, or rub the cursor over its head for a headpat; 7 palettes, 4 sizes, synthesized sounds, light/dark theme sync, and a playable web preview.

Replay sessions on a playable timeline with per-step token usage, audit sensitive operations, estimate cost, view fork lineage, compare sessions, and export standalone HTML replays.

Persona engine for DeepSeek Harness: one config.json drives the AI's self-name (per model, per tier), how it addresses you, relationship stance, character text, per-item work contracts with individual toggles, thinking-chain language, appearance and reply tone (both opt-in and per-model overridable), and long-term memory gated by human confirmation - the AI can only append proposed candidates, a human confirm line flips them into effect, and the log is physically append-only and replayable. Confirmed entries are injected as a quoted data-not-instructions block; entries tagged with a kind can instead be routed to arbitrary files on confirmation and never reach the prompt. Three prompt sections re-evaluate from disk on every assembly, so edits apply on the next step with no restart. Persona presets are shareable JSON snapshots (imports and exports SillyTavern character cards, v2/v3 read, v2 written) and deliberately never carry memory. Zero runtime dependencies, no network, no shell; any error degrades to an empty section rather than a broken session. Ships blank - installing changes nothing until you write something.

Laya typed decisions (`noul` yes/no, `choice`, `score`) as a first-class Cordis service (`ctx.laya`) plus the model-visible `laya_ask` and `laya_plan` tools. The plugin itself installs and downloads nothing: it is a client of a `laya-mcp serve` sidecar you start yourself, and it reports that the sidecar is not running instead of failing obscurely at the first tool call.

Select text in the DSH web UI to stream a translation and an explanation of what it means in that specific context, read against the conversation it came from; the panel keeps a replayable history and can be promoted into a full session.

Listen to Ximalaya podcasts and audiobooks inside the DSH web UI: search albums, browse paginated tracks, and play through a now-playing bar (prev/play/next, seek, volume, rate). After QR login the "Mine" page lists your liked tracks (click to play), subscribed albums with latest-episode hints, and followed anchors with their public albums; the host relays audio streams with Range support and registers an `ximalaya_play` tool so the agent can search and play on request.

Full-screen video boot splash — play an mp4 at startup, skip with a click, import your own clip, and non-playable codecs are transcoded to H.264 automatically.

A neon pixel theme for DeepSeek Harness desktop with playable air-battle, pig-sling and fruit-slicing backgrounds, mouse controls and a day/night switch for the sling game.

Hides tool schemas from the model's prefill and reveals one tool at a time through a decision model API. Requires a running decision-model API endpoint and pnpm on PATH.

Read-only clinic for DSH session logs: 22 named diagnostics (token waste, retry and failure loops, context bloat, cross-session amnesia, idle MCP servers and skills), each reported with a replayable evidence locator such as session.v4.jsonl.zstd#L101-L110, and each finding turned into an executable prescription — a profile-patch YAML snippet or a command with current to suggested values, risk, verification and source — plus a feedback command that retires a prescription you reject, after which the report states it was removed by your feedback. Zero dependencies, default zero egress, session files are never written.

Adds a manim_card tool that renders Manim Python into a playable animation card in the chat stream; the code runs in the browser on a bundled runtime, so no local Python is needed, and video export goes through the local Manim CLI when installed or frame by frame in the browser otherwise.

DeepSeek Harness approval guardian: a reviewer model decides permission requests before they reach you, with a System One decision-API backend (TypeSafe Jev, SiliconFlow, self-hosted Laya), a configurable low-confidence policy and an allow-rule whitelist.

Mail list/search via himalaya / notmuch (no send).

/repro exports a minimal, secret-scrubbed, replayable problem bundle: the session log, failed commands, and Git diff.