Plugins

Zero-key market data MCP server (dsh bundle): 5 MCP tools for A-share/HK/US quotes, sector rankings, futures, money flow and 7x24 news, with a three-tier upstream fallback.

Merge every locally signed-in WorkBuddy account into DeepSeek Harness as one auto-failing-over model pool: multi-account rotation, 429 cooldowns, live credits, a model catalog with per-model credit multipliers and free/night-discount tags, QR-code login for 13 vendors, and aggregate free-tier channels. Accounts persist across restarts.

在 dsh 里装上这个插件即可，无需登录、注册或填 API Key，就能使用包括 Muse Spark 1.3、MiMo V2.6 在内的前沿模型——完全免费，不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API key — the frontier models are just there, Muse Spark 1.3 and MiMo V2.6 among them. Completely free, with no usage cap.

Dispatch work to DSH agents from Claude Code or Codex: native subagent progress, in-host worker sessions with per-tier presets, and a multimodal bridge for vision and image generation.

Proactive associative memory for DSH: zero-prompt recall injected at a fixed boundary, three-layer auto-consolidation, skill crystallization, and Astra-style context management - handoff ledgers and a PLAN whiteboard that survive context-window switches, water-level sensing with an auto-detected model window. Local Markdown storage, model-agnostic, zero deps; lexical 0GB floor with optional semantic tiers.

Three-tier memory control plane for DeepSeek Harness: persistent runtime context, searchable project documents, pluggable long-term memory, smart routing, supervised agent workflows, WebUI, and headless tools.

Asset governance console for self-hosted AI infrastructure. Registers MCP servers, skills, agents, AI apps, model channels, SaaS connectors and NAS storage in one ledger with a lifecycle state machine, per-tool RBAC permission points, department x agent x skill cost attribution, dependency topology and an append-only audit trail. Self-hosted, MIT, no hosted tier and no cloud contract.

A Claude-persona desktop pet pinned to the bottom-right corner of the DSH web GUI — shows balance, today's usage, peak/off-peak pricing and per-turn cost, with editable bubbles (text, balance, usage, pricing, images, random lines and weighted choices), per-line styles and fonts, a floating quick editor, sound effects, and custom character/audio. Ships animated roles that breathe and blink, plus a balance-alert variant that switches to a crying face when the balance falls below the alert line; 256/384 animation tiers and an optional 512 HD pack installed from the resource manager. 63 Claude-flavored lines. A derivative of MeteorNOX/DeepSeek-Balance-Whale-Widget (MIT).

Pins the model's reasoning language to a tier picked in Settings - General - Thinking language (Simplified Chinese, or a default tier that only constrains replies and leaves reasoning to the model), by injecting a system-prompt section at order -1000000 and re-hoisting it to first place at the end of the assemble waterfall, plus an optional per-turn copy placed ahead of each user message for models that follow static sections weakly.

Puts the Markdown documents, source files, images and video that already exist anywhere in the session workspace into the DSH sidebar, ranked by relevance to the current conversation. The default lexical mode matches your recent messages against title, summary and body locally with IDF weighting (BM25) — no model calls, no network, nothing leaves the machine. The list comes from scanning the workspace rather than from a recent-files list, so restarting DSH or opening a new session does not empty it. Results are organised into a three-tier reading order (primary / supporting / related) with a verifiable reason on every row — which field matched, which document cites this one — and one click opens an in-place preview: Markdown as rendered text, relative-path images actually drawn, source code with the host's own syntax highlighting, video streamed over HTTP Range. The same Context Pack is exposed to the agent as a knit_docs tool. Later versions added read evidence and lifecycle tracking (attribution is frozen at the moment of reading), pin / exclude (exclusion applies before retrieval, and is undoable), relation evidence (references / imports / tests, with line numbers) and an incremental index reused across restarts. Since v0.23 there is also an optional hybrid mode: it is off by default and only runs once you configure an Ollama or OpenAI-compatible embedding provider yourself. When it runs, what leaves the machine is your query text plus each candidate file's title and summary — a Markdown summary is an excerpt of the body's first paragraph (at most 60 characters), a code file's summary is its workspace-relative directory, and an image or video contributes its file name; full bodies and absolute paths are not sent, and a loopback provider keeps everything on this machine. Since v0.24 it also associates uncommitted changes: a read-only git status (fixed argv, no shell, whitelisted child-process environment) combined with the same reference / import / test / mention evidence shows which files a change is likely to drag along, and it is off by default. The benefit of semantic retrieval still awaits a real-provider run, so no recall improvement is claimed.

One-command backup, restore & migration for all DSH user data (~/.dsh) with a visual Settings panel: scheduled auto-backup with tiered retention, sha256 verification, hardened restore with auto-rollback and a result receipt, typed sub-backups (--types), credential redaction into a local vault, private-repo GitHub sync for moving to a new machine, and an out-of-process rescue console (zero-dependency rescue web UI in the backup dir) that works even when the host won't boot. /backup doctor checks and repairs corrupt session logs from a known-good backup and quarantines them before rotation; migrate-check statically predicts which sessions a host upgrade would break (frozen event-list rules, filename/generation mismatch, hardlink support, legacy credentials); the credential sentinel preserves pre-migration copies before the host rewrites them; smart backup snapshots before host upgrades; /backup check-update and /backup update report and install new releases (snapshotting first). Restore handles attachment hard links and arbitrarily large archives (streamed listing). Current npm version 0.13.4; spans DSH 0.1.1-rc.2 through 0.2.0-rc.1, including 0.2.0's install-time peer gate. macOS/Linux/Windows.

Brings WorkBuddy desktop-app models into DeepSeek Harness: the China and international builds each appear as their own group in the model picker, with the reasoning tiers the upstream declares, per-reply credit cost and remaining balance, per-account model visibility, and image input where supported.

MCP server manager (stdio / streamable-http): per-working-directory project and global config tiers; project-level MCP collapsed into 4 atomic tools through a middleware layer by default (`middleware: all` folds in global servers too, hot-switchable in the settings page); workspace isolation prevents cross-project interference; configs keep `${ENV}` references only, with no plaintext secrets on disk; runtime registration API for other plugins to inject MCP servers.

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.

Detects prompt-injection, jailbreak, and secret-leak patterns on the agent/pre-step, tools/pre-execute, and tools/post-execute seams with allow/ask/block tiers, sanitized defend/detection audit events, a defend_report tool, and a destructive-delete command guard.

LLM-assisted auto approval with countdown fallback for the Auto permission preset: static rules, risk tiers, breaker and file audit.

Usage tiers and shareable stats cards: 🐟→🐳 ranks by monthly token burn with a locally-computed estimated percentile, precise pricing for 46 models across 6 vendors (including size-tiered Chinese pricing), backfill of pre-install sessions, and old-vs-new comparison across the 2026-08-17 rate change. Local-only, no telemetry.

Rust-based POSIX bash executor for the Windows platform — gives DSH a native bash with no WSL or MSYS compatibility layer, honors DSH's three-tier permission policy, and provides persistent-terminal tools.

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.

Seven agent-preset modes for DeepSeek Harness, each with a fixed tool boundary, per-mode model routing and fast/deep omd_task delegation tiers, plus a plan-to-execute loop (/mode, /start-work, goal-backed /ulw) and a settings page that remaps models per mode and per tier.

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+.

Pins every request to the native max-effort tier and adds a per-model/per-mode reasoning-depth panel with a 0-20 capability slider, isolated Velocity/Apex profiles, and offline reasoning toolboxes. Distributed as an obfuscated build; commercial use allowed, reverse engineering closed.

Free vision bridge for text-only models: image understanding, OCR, UI and debug analysis via free-tier providers (Qwen3-VL-Flash, Doubao, DeepSeek-OCR) with a settings GUI.

Free OpenCode Zen models in DeepSeek Harness, anonymous by default with an optional Zen key. The catalogue is read from Zen's official model list at runtime, and anonymous requests are adapted to the free tier's request checks, documented with dated measurements.