Plugins

Modular AI research/engineering skill pack, installable via dsh plugin add — registers 27 skills (security audit, paper reading/writing, figure drawing, dev workflow, storage analysis) on ctx.skills.

Cross-session memory with zero-token deterministic recall (bigram-Jaccard fused with FTS5 BM25 via RRF, Chinese-aware tokenization), reinforcement that strengthens repeated observations, and procedures crystallized into SKILL.md files refined in place through an active-stale-archived lifecycle with reversible archive and rollback. Ingest and disposal each expose a user-chosen autonomy tier; conflicts and high-importance writes always go to human review, and nothing is ever auto-deleted. Requires Node 22.5+.

Analyze usage across models, providers, workspaces, and time ranges — including custom start/end dates over the full retained history — with live cache-hit visibility and separate model and provider breakdowns. Includes hourly single-day trends, daily cross-day trends, a 53-week GitHub-style heatmap, usage streaks, model and workspace Token-share donut charts with Token, cost, and percentage details, independent combinable filters, cache-efficiency trends, DeepSeek account balance, workspace aliases, CSV export, and a durable ledger that preserves successfully flushed usage after session deletion — and after a workspace is deleted, where its recorded usage is summed into one Deleted row — and rebuilds incrementally — on restart, unchanged sessions are applied from the ledger without re-reading their logs (a per-session log revision is the change signal), and only changed/new events are folded. The workspace registry is kept fresh automatically through the DSH domain/changed probe: only added/removed workspaces are reprocessed incrementally, unchanged workspaces reuse their computed aggregates and ledger with zero rescan, and usage is strictly limited to DSH-registered workspaces (unregistered cwds are ignored). After a scan, lightweight revision status fetches full history only when the Host/data state changes and exposes non-sensitive sync health (revision reuse, rereads, ledger recovery, and failures). Token accounting is explicit: input is fresh (cache read/write and reasoning are separate buckets) and zero-usage replays never overwrite recorded usage. Chinese uses local time, English uses UTC; API access is loopback-only.

Multi-engine text-to-image generation (OpenAI Images and Zhipu CogView presets) with per-session quota tracking, engine failover, credential-safe config, and a result card with regenerate.

XiaoTangYuan, an in-game AI companion for single-player games: understands the game screen, chats by text or push-to-talk, and speaks replies aloud; Stardew Valley is supported, with Don't Starve Together and Oxygen Not Included integrations in progress.

Plugin marketplace in Settings that targets any local profile, not just the running one: install, update and remove across web, headless and other profiles, with install-script warnings, an installed view that also covers non-catalog profile dependencies, and an automatic DSH restart when the current profile changes.

Continual self-refinement loop: persistent memory, periodic review-and-refine, cross-session shared knowledge, and automatic rollback through a model-callable harness_refine tool.

Chrome side panel that embeds the dsh web UI and lets the agent read the current page, capture HTTP traffic, and drive the browser.

Injects one fixed system-prompt section so the agent thinks and replies in Chinese, regardless of the user's language.

Usage and cost statistics with peak/off-peak pricing after the 2026-08-17 rate change, a floating summary panel, per-session breakdowns, and a day/week/month/year/all usage heatmap.

Composer input toolkit: arrow-key session history (first/last-line gated, IME and command-menu safe), an instructions viewer/editor for global and project AGENTS.md/CLAUDE.md, and a prompt library of 780 Chinese prompts (Cherry Studio agents-zh corpus, AGPL-3.0) inserted into the composer in one click. Ships on npm as `dsh-composer-tools`.

Codex-style drag-drop: drag any file into the DSH web GUI, it lands in ~/.dsh-dropbox, and the path is inserted into the composer as a whole blue chip.

Bilibili video analysis: metadata, transcript (ASR fallback via Bijian/sherpa-onnx/whisper.cpp), comments, danmaku, and sharp keyframes with optional local vision descriptions.

Token, cost, and balance dashboard under the composer: live indicator plus zero-dependency SVG charts for per-turn usage, estimated cost, and DeepSeek account balance.

Right-edge user-message navigator with an adaptive block layout that fits the available height, plus hover previews, keyboard controls, and click-to-jump navigation.

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.

Paper reading workbench: library + PDF reader + tutor-style companion. Answers cite clickable page numbers that jump back to the PDF and flash the exact passage; select-to-ask without leaving the paper; bundled strict-tutor preset (reading plan, section coaching, graded quizzes tracked across sessions); in-place original ↔ Chinese toggle (babeldoc, zero preinstall). Local transcription and search — documents never leave your machine.

Tool-call discipline for DeepSeek Harness: injects a static behavior-constraint prompt section, drops the parallel pool to 1 after a tool failure (restoring after 3 consecutive successes), nudges the model to stop retrying after repeated failures, verifies writes with .bak rollback, gates high-risk commands, and tracks compliance stats via /behavior-status.

Layered token-optimization pipeline for DeepSeek Harness: output ladder, MCP lazy loading,compaction driver, cache-hit reporting. Built on real DSH plugin APIs; ~40-60% input saved in long sessions.

Evidence-first attention supervision for DeepSeek Harness with C0–C3 policy, deduplicated Inbox alerts, quiet hours, and redacted outcome records.

Self-contained literature library for DSH with optional Zotero export: full-text PDF fetching (Unpaywall/arXiv/custom URL templates), loose Crossref-backed search, APA/GB-T 7714/MLA/Chicago citation generation, a side-panel PDF reader with highlights and notes, and AI-assisted Q&A/translation steered into the current conversation.

Server card dashboard for DSH: per-host status, CPU, memory, disk and latency, with one-click xterm.js SSH terminals; imports ~/.ssh/config and mounts as a better-sidebar tab or a standalone right drawer.

For the DeepSeek Harness native vision model deepseek-v4-flash-vision-exp, raises image admission limits to 32 MiB / 8192 px / 600 images and adds a highres_read tool that tiles large images, then returns the whole image plus 800x800 tiles through the host read_image tool.

Programming Mode agent preset bundle: the standard coding agent with its persona mandating the Superpowers discipline (skills first, brainstorming before building, plans before code, TDD, systematic debugging, verification before completion) plus an always-on Ponytail code-volume discipline (7-rung minimal-code ladder, default full); ships all twenty skills bundled (14 superpowers + 6 ponytail) and mechanically injects the full using-superpowers skill into every new session's first request.