Plugins

Completion sound notifications for DeepSeek Harness: plays an embedded chime when a session's output finishes, with a Notifications row in General settings. Settings persist in browser localStorage — a standalone web plugin, no Host settings dependency.

Animated DeepSeek blue-whale desktop pet floating on the DSH Web page: draggable with speech bubbles and mood poses, a pet_say tool for the model to make it talk, and turn-start/end thinking and celebrating states.

A draggable mascot invites you to play Xiangqi (Chinese chess) while the AI thinks, with a built-in engine, move-record export, and optional multi-model commentary.

Project-scoped persistent key-value memories with set/get/list/delete/search model tools and JSON-file storage.

Opt-in pixel-neon arcade skin with Fusion Pixel fonts, animated whale-girl icons, custom crosshair and draggable detail windows; preserves light/dark/system themes. v0.3.2 includes local display-name controls, mobile menu improvements and an optional source patch for the arcade home screen, transitions and telemetry HUD.

Floating API-balance badge for DeepSeek Harness: live balance for DeepSeek, Moonshot, OpenAI and custom endpoints; draggable, resizable, frosted-glass look, with text color adapting to the content beneath.

Google AI Studio-style floating preview for the Web UI: a draggable/resizable iframe preview with aspect-ratio and resolution presets, plus a VS Code-style code window with direct editing, confirmed writes, and session-log recording.

Floating community panel for the Web UI: 24h hot topics and hot plugins TOP10 with a draggable, always-on-top button.

Persistent long-term persona memory for DeepSeek Harness: MEMORY.md/USER.md storage, memory + memory_search tools, per-request context injection, and secret scanning. Ported in spirit from pi-hermes-memory.

A draggable, resizable floating panel that simulates an air conditioner, light, curtain, and air purifier, with an agent-callable control tool. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.

GitLab issue kanban for DSH, driven by the glab CLI. Every label-backed list of a project issue board becomes a column; dragging a card swaps its status label. Adds a sidebar panel with a project picker, filters and a card detail drawer (Markdown description, comments and activity, comment, close, reopen), plus a session View tab that resolves the workspace origin remote, and one glab_kanban agent tool over the same operations.

A proactive coding agent as a DSH bundle: it listens to GitHub issues and comments, and optionally Feishu, DingTalk and WeCom, turns requests from authorized people into tickets, runs each ticket in its own DSH session and git worktree, and replies in the original thread with a branch or pull request.

Drone coverage route planning panel for DeepSeek Harness: draw point, line, area, slope or cylinder targets, set route presets (altitude mode, speed, yaw mode, stay, end action, range cut), then generate boustrophedon scans, route-aligned lines or cylindrical spirals, with a waypoint table, risk checks and per-waypoint actions.

Find existing GEO projects and reports and review sample coverage, competitors, and citations with read-only Zhixing Ark authorization.

Checks the agent’s own claims against the tool output saved in the same session: each sentence carrying a path, a number with a unit, inline code or a URL is graded supported, partial or no evidence, with the anchors it could not find listed. Coverage shows on the composer pill. Heuristic and local — no LLM call, no receipt log, nothing uploaded.

Colored tags for sessions, shown on the sidebar row and in the conversation header, with preset colors, pin-to-sidebar, hover-remove and drag-to-reorder; tags persist in a Host-side storage domain and survive restarts.

Tactile warm-paper skin for DeepSeek Harness: seven built-in palettes plus AI-designed themes, tight typography, spring sidebar motion, switchable streaming-text reveal and comet caret, per-workspace icons, and draggable sidebar entries with a "More" tray.

Adds a session tab bar under the DSH conversation header: open sessions appear in order with workspace/title labels and status dots; click to switch, use the hover × or the right-click menu to close, rename, close others or close all; the tab list persists in localStorage. Plugin only; it does not change DSH core or stop background work.

Connects personal WeChat direct messages as a dsh-superagent Source, so private-chat text becomes a trigger for the orchestrator, accepted tasks run as work orders, and results are replied to the same chat; group and non-text messages are dropped. Only WeChat users on the superagent policy's allow-list are assessed; the list is empty by default, so every DM is ignored until you add your own ilink_user_id.

Before you say yes to an agent, ask how the code works: answers come with an interactive UML diagram (8 types) linked to the text, so pointing at a paragraph highlights it in the diagram and clicking a step jumps to its paragraph. Code quotes are checked against the files.

Project board in the sidebar: a self-contained HTML Kanban and Gantt page served over a same-origin route, drag-and-drop columns, due-date timeline, JSON import/export, data kept in local storage.

Skill manager for DSH: drag rows into the order you want, pin any row straight to position 1, and rename a skill inline. Applying rewrites each directory name and the name field in SKILL.md. A skill that appears later is numbered and cleaned up automatically, and a description past 20 Chinese characters gets rewritten. Every change clears the host skill cache and notifies each session, so the slash menu follows at once without restarting DSH. Local-only and offline. Its HTTP routes have no DSH access check, so a page open in your browser could trigger a rename or reorder of your skill folders.

Session tree with tag chips, group-by-tag, drag-to-move across workspaces, workspace ordering, favourite sessions, a four-state filter, and a delete action that backs sessions up first.

Grows a conversation into an editable card tree where each card is one self-contained question and answer, with child cards, same-level branches and in-card follow-ups, and an overview for dragging, re-hanging, cross-referencing or pruning the graph.