Plugins

Compact 56px app navigation rail with a separate conversation list and sidebar collapse control for the DSH web UI; also hides embedding, rerank and image/video generation models from the chat model menu and /model.

Sidebar panel for a weekly, human-curated Chinese ranking of the DSH ecosystem: 31 tracked projects (17 installable plugins plus tools/skills) ranked by live GitHub stars with week-over-week deltas, and a real-call availability survey of the "free" models. Sponsor slots are labelled and sit outside the ranking.

Generate images from chat by calling a fleet of local or LAN ComfyUI workers, with workflow templates, worker health checks and queue control.

Watches the foreground browser window, recognizes video courses, classifies them by subject, fetches Bilibili metadata/danmaku/comments, and generates Markdown + Mermaid knowledge maps and Word study manuals.

Shows what each assistant turn cost under every reply, priced from the token usage of that turn at the official peak and off-peak CNY rates, together with the account and bonus balances; when a turn cannot be priced it states the reason instead of guessing.

Favourite any conversation from the sidebar row menu or an inline star, copy its id plus the transcript absolute path in one click, and permanently delete a single conversation behind a confirm — the sidebar row drops at once. A settings card batch-archives sessions idle past a configurable hour threshold with one-click undo, exports the candidates as a tab-separated txt carrying name, id, workspace, idle time and last activity, and batch-deletes archived conversations that are not favourites. Two view-options rows narrow the sidebar to favourites that are archived and to those that are not. Every addition is additive — three official slots plus one cloned filter row, editing no official file.

Vision understanding + image generation/editing for DeepSeek Harness (read images with Qwen/OpenAI/Anthropic/Gemini, generate and edit with Qwen-Image). A rework of liustack/modlens (MIT). Remove @liustack/modlens first — both register the same vision providers and image interception hooks, so co-installation breaks image reading.

TangSan mascot appearance layer for the DeepSeek Harness web GUI: wears a generated TangSan portrait as the GUI background (surface-token fade with opacity, blur, scrim and position controls), seats the matching square avatar as the sidebar and conversation brand mark, and edits it all from a Settings section; the host serves the artwork and config locally under /api/tangsan, and the config lives at $DSH_HOME/tangsan.json.

Companheiro de anime para Windows: exibe saldo DeepSeek e cotas Codex, com OAuth multi-conta do Codex, gestão de cotas/modelos por conta, modo compacto e configurações salvas automaticamente. A troca de conta no painel é opcional. O warm-up diário exige autorização explícita separada. Gatilhos startup/5h-full multi-conta ainda não estão disponíveis.

Let a DSH model inspect and operate Windows applications through an installed Codex Computer Use runtime, with per-session app approval.

PostgreSQL expertise for DeepSeek Harness: 11 curated pg-aiguide skills from Tiger Data registered natively, plus the pg-aiguide MCP server for version-exact manual search.

DSH 生图+生视频整合插件：统一多供应商目录（OpenAI 兼容 images/chat 网关、DashScope 通义万相、火山 Seedance、Google Veo、Sora 风格 /v1/videos 异步任务含 input_reference 图生视频），提供 image_generate / generate_video / animate_image 工具，对话内联显示、图片与视频画廊、统一设置面板。Fork 合并自 dsh-image-generation 与 dsh-video-gen

A fork of zhu1090093659/dsh-pet (Apache-2.0), adding per-conversation companions with whale-girl, MIKU and business blue whale skins, model-themed colors, token-rate animation, completion reminders and shared feeding and affinity.

Adds a Hindsight configuration and status dashboard to the DSH sidebar for connection, model, embedding and bank settings, with health checks and ownership-checked local service controls; memory tools are provided by the separate official integration. Hindsight itself is a separate service and is not installed by this plugin.

Exposes the Paper-gogo evidence-first paper workflow as native tools: all 18 phases and the seven G0-G6 gates become callable, gate results are normalised and advancement is refused while any gate fails, and claims are audited against the four evidence states SUPPORTED, INFERRED, VERIFY and MISSING. Ships the workflow text and 57 skills in its own directory, so it installs without a separate Paper-gogo package.

Enumerate and control Windows top-level windows from WSL: activate, minimize, maximize, restore, move, resize, set always-on-top, close.

Windows UI Automation from WSL: enumerate top-level windows, read a bounded element tree, and wait for or trace the path to a named element.

Windows desktop automation for dsh: 14 tools that enumerate apps and windows, capture window screenshots that still work when the window is covered by other windows, read the accessibility tree, and perform real input — click, type text (CJK included), press key combinations, scroll and drag. Loads the official computer-use runtime (@oai/sky) already installed on the machine, discovering it by enumerating vendor directories, so vendor upgrades that change hash directories do not break it.

DeepSeek Harness 插件合集（整合包）：按人群一键装齐整套插件。Curated persona plugin bundles for DeepSeek Harness.

One-tap connect Agnes language, image and video models into DSH (off by default), with image recognition and thinking level filled in automatically. After signing in to the Agnes console (re-login uses AGNES_PASSWORD), you can see the quota windows (5-hour requests, weekly requests, daily images, video), account-wide cumulative usage, and a usage pill beside the input box. AgnesCode desktop models can be connected too (Windows only), re-reading the desktop app login when the token expires, with rate-limit hits backing off on their own.

Shows whether the DeepSeek API is currently billing at the off-peak or peak rate in the sidebar brand row, with the Chinese statutory-holiday calendar built in and zh/en copy.

Read-only Skill inspector for DeepSeek Harness. It shows which Skills a conversation actually loaded — from observed loading evidence only — and which Skills the current environment can discover, and gives each Skill a detail page whose left column switches between the framework, the evidence, validation, evaluation, step evidence, run logic, the before/after of a modification when there is one, the definition facts and the raw SKILL.md. Declaration and run stay apart: the body is split into structure, declared workflow and progressive disclosure, the run logic is read from local receipts, and every declared step carries its own evidence — no observed evidence is reported as exactly that, not as executed or not executed, and the run's instruction hash sits beside the current definition as match, mismatch or unavailable. The evidence view gives each subject one of three states (declared, observed, unavailable), walks definition → load → use → outcome, names what each stage cannot reach, declines to attribute an outcome to the Skill, and can copy the snapshot out as byte-stable JSON. Validation checks the Agent Skills open standard and the platform profiles separately, each finding on a rule id and labelled standard compliance or platform compatibility. Evaluation keeps repeatable cases with the conditions each run was recorded under and the four stages of runtime evidence, comparing two runs of one case item by item without scoring them. The trigger boundary keeps should-trigger and should-not-trigger prompts, has you run them in a real new session, then records what was observed — whether a load record appeared and whether it was model-invoked or user-explicit; a typed /name is not a natural trigger, and an absent load record is reported as not observed, not as not triggered. The same projections are exposed to the agent as a read-only skill_review tool, which returns a structured review packet and says what it omitted when the packet will not fit whole. The inspector, the evidence chain and these reviews are computed on this machine from the definition and the session receipts; the Chinese reading version sends the body through the host's model when you ask for it, keeping code fences, paths and frontmatter verbatim and storing the result on this machine. A Skill can also be cloned into a local copy. It does not write back to the source Skill and does not run its scripts.

Session cost in rubles in the composer dock — the tariff in effect with its countdown, the API cost at that tariff, the CBR rate, the total including subagents, cache savings, and DeepSeek Platform balances.

Cut LLM costs: a Jev (System One) classifier vets each turn against a declarative policy and recommends the subagent model route, so routine work runs at flash rates while the main model keeps the hard parts; explicit /route trigger, runtime delegation-capability gate, credential redaction, and offline policy replay over the verdict log.