Plugins

Enterprise AI resource platform as a 22-plugin cordis bundle: org/account/role IAM, OIDC provider and App SSO, MCP server management with the platform itself exposed as an MCP server, NAS asset storage, skill hub, agent and app registries, SaaS connector management, enterprise portal data channel, department agent workbench panel, DingTalk group bridging for message and approval push, one-click remote dsh enrollment with tool proxying, usage metering, wallet billing, model gateway, audit with alerts, self-update, and a reviewed third-party plugin marketplace — 70+ ops tools become callable by the dsh agent.

A 60-frame theme-aware monochrome whale-dive animation with reactive water for the DSH Web turn status: embedded assets, seamless closed-loop playback, a reduced-motion PNG fallback, and lifecycle-clean removal.

Git-style conversation branching that mirrors how real programmers collaborate: a secretary root branch dispatches tasks to child branches with their own contexts, merging back only compressed summaries via squash or rebase; ships /branch commands, a branch tab, branch-to-branch messaging, and a governance baseline pairing each session branch with a git worktree.

Context Assembler plugin for DeepSeek Harness: topic-block context assembly, water-pressure topic splitting, tool trace/rewrite, reality recall injection and handoff planning.

A slim port of Headroom: compresses tool output, cuts up to 20% of context, without affecting the model's context cache or agent performance.

Chinese localization for DeepSeek Harness: translates all system prompts, tool descriptions, and runtime contexts into Chinese at runtime via the system-prompt/assemble waterfall (1788 entries, zero source modification, uninstall restores English).

Evidence-first conclusion gate that cross-checks first-hand claims against a durable per-session tool-activity index, blocks unverified guesses and tracks blocked-versus-converted metrics.

Browser notifications and a beep when a session turn ends, the agent asks a question or approval is pending; clicking the notification jumps to the session; remote trigger endpoint and SSE channel (npm: `bsfeng-dsh-notify`).

Interactive whale-girl desktop pet for DeepSeek Harness: live balance/usage/context occupancy, slingshot toss, easter-egg bubbles, eco mode. 346KB package, zero extra processes.

Vault button beside the DSH composer: save keys, servers, phone numbers, addresses and URLs, auto-collected from conversations, password-protected with WebDAV backup; click to insert, Ctrl+click to send.

Web Audio lifecycle cues for DeepSeek Harness: ten configurable events, three low-noise defaults, per-event custom sounds, and no shipped audio assets.

Load Claude Code assets into DSH at runtime — skills, slash commands, rules, permission rules (deny/ask/allow), subagents and hooks from .claude/ plus Claude Code plugins (plugin.json / marketplace / skills / agents / MCP), via a memory-IR parse layer that keeps .claude as the source of truth with no writes back. Six installable packages: loader, skills, permissions, agents, hooks, mcp.

12 tested engineering skills (git commit, code review, test-first, plugin dev, debugging, changelog, PR review, task breakdown, and more): every SKILL.md passes a format validator aligned with the official skill rules plus an isolated load smoke test.

Startup splash for DeepSeek Harness: an animated avatar with aurora, stardust, pointer parallax and a module-assembly entrance, plus the same image as the main-UI wallpaper; both images can be swapped from a settings page that previews the animation live.

Tripo 3D skills for DeepSeek Harness — generate engine-ready 3D assets from a text prompt or a reference image via tripo-cli, with texturing, rigging, retopology and format conversion in one pass.

DSH Desktop enhancement plugin: it puts verifiable facts on the table. Constraints are injected only when relevant, so idle chat costs no extra tokens; injection is layered (session-stable text in the system prompt, volatile content as a tail snapshot) and the prompt cache stays valid. Discipline: each turn is classified (question / research / discussion / diagnosis / execution) with boundaries enforced - no file edits on question turns, no unsolicited fixes on diagnosis turns; citing code lines not opened this session, negative claims about unseen symbols, requirements that do not match the deliverable, and dressing up “I did not check” as “waiting for your decision” are all called out; replies are written for humans - a code, field name or id must be explained in plain words the first time it appears. Memory and knowledge: 75%/90% context warnings plus an exported session recap (say "continue" in a new session); project knowledge accumulated across sessions by working directory, auto-captured and deletable by id. Inspectable artifacts: task contract / change record (auto-logged, unverified items flagged at delivery) / hypothesis record (conclusions must pass an evidence gate) / design decisions. Dashboard and personas: routing, trigger hits and block assembly are written to lume-metrics.jsonl and queryable via lume_metrics; personas are distilled from chat logs, novels, scripts or character sheets, with long-term and temporary memory, a memory star map and persona card export/import.

Switch the language of the agent's thinking/reasoning process (chain-of-thought): 简体中文, English, Русский, Français, Deutsch, Español, 日本語, 한국어 and more, or auto-follow the system locale. Adds a Settings → General row, a system-prompt instruction (per-prompt assembly, applies to new sessions), and the /thinking-language command.

Unified Godot asset and project-goal management for DSH: godot_* model tools, a /gap REST surface, and a better-sidebar UI with asset manager, scene-tree viewer, and Git panel.

DeepSeek Harness Web plugin — PDF text-layer and scanned-PDF OCR, Office docx/xlsx/pptx to Markdown, TIFF/epub and long-document index cards.

Bounded, cell-addressed `doc_read`/`doc_write` for xlsx / pdf / docx / pptx / ipynb, plus an MCP server and CLI.

Polishes the draft in the composer into a clearer, better-structured prompt from the ✨ button or Alt+O; streams the result live over SSE with reasoning shown first, and follows the session's default model with zero configuration or any OpenAI-compatible endpoint.

Matt Pocock's 25 skills fully translated to Chinese (technical terms kept in English with glosses).

Bounded, cell-addressed `doc_read`/`doc_write` for xlsx / pdf / docx / pptx / ipynb, plus an MCP server and CLI.

A Witch on the Holy Night (Mahoutsukai no Yoru) visual-novel theme for the DeepSeek Harness Web UI, with original-style menus, three characters with personas, day and night playlists, and an ending sequence. The repo ships no assets of its own; the 380 files / 263 MiB it drives (136 sprites, 208 UI parts, 13 backgrounds, 23 audio) must be recovered from your own copy of the original game, and the screen stays largely empty until then. The 23 audio files are a header strip away from working OGG; the 357 images use proprietary .mzp/.cbg formats, so the per-file mapping guide offers two routes and one of them expects you to supply your own decoder.