Plugins

Visual multi-agent DAG workflows for DeepSeek Harness with live execution, restart recovery, and Feishu integration.

Adds an optimize button beside the composer that analyzes and rewrites the prompt draft with SSE streaming, conversation-context awareness (template vs intent-polish strategies), an edit-continuity memory chain, undo, and model routing with fallback and a connectivity test.

aginxbrowser plugin for DeepSeek Harness: fetch JS-rendered / Cloudflare-protected pages as clean markdown, aggregated web search (Baidu/Sogou/WeChat and 11 more), and indexed interactive sessions (click / type / scroll / eval / screenshot) against a host

One-click prompt enhancement in the composer — rewrites the draft into a structured prompt (role and goal, executable steps, output format, acceptance criteria, edge cases) with before/after preview, fill-back, and undo. Never changes intent, never fabricates requirements.

Local-first code knowledge graph for coding agents: 10 MCP tools for layered context compression, DAG planning, incremental AST indexing, a cross-session skill flywheel, diagnostics, and graph artifact import/export.

Gives a text-only model image understanding via a vision-language model, exposed as a `describe_image` tool.

Coding-subscription OAuth for SuperGrok/Grok Build, ChatGPT Plus Codex, Kimi Code and Claude Code, plus an opt-in loopback OpenAI/Anthropic gateway: local login without pasting tokens, AUTH refresh-and-retry, grok-4.6 xhigh, and API-key route coexistence.

Right-panel system for the dsh web GUI: a pixel-faithful AionUi Explorer + Preview (file tree, diff view, preview).

DeepSeek Harness plugin for dsh-baton: adding it is the whole install. Switched on, it registers the dsh-baton MCP server for Claude Code and Codex, then takes the baton over a local, owner-only control socket, runs it in a specific session through the ho

Injects a fixed system-reminder at the first step of every turn asking the model to reason and answer in Simplified Chinese, while code, commands, file paths, and identifiers stay in English.

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.

Views and builds CAD models (STL, OBJ, STEP, IGES, DXF, SVG) via cad_view/cad_info and OCCT parametric modeling tools, with an interactive viewer card and a resident CAD panel in the Web UI.

Bridges WeChat private chats into DeepSeek Harness as an in-process plugin over Tencent's iLink bot protocol, with two-way text, image, file and voice messaging, in-WeChat slash commands resolved through the native command registry, WeChat-side handling of approvals and agent questions, and proactive push through a send_wechat tool.

Edit a sent message to rewind and branch the conversation from that turn, switch versions with a counter under the bubble, and browse every branch in a turn-level version tree.

DSH（DeepSeek Harness）插件：把 ego-lite 浏览器（给 AI Agent 用的 Chromium）接入 HARNESS——13 个结构化 ego_* 工具（文本语义快照、语义定位点击、表单填充、截图、CDP 控制、任务空间隔离），内置 ego 运行时，Linux + Chrome 开箱即用，无需克隆官方仓库或手动构建。

Project memory for dsh with native task-system integration: session todos and file reads auto-condense into cross-session task records (task↔file links), so workflows are switchable and resumable without re-contextualizing. Read-time memory (no upfront scan): extracts code symbols, summarizes docs, and cross-links docs↔symbols; optional TS semantic enhancement (auto on when TS installed). Experience notes auto-dedup. Every recall includes a traceable citation. Pure JS, single dep, works out of the box, silent operation, zero overhead.

Windows tray desktop shell: auto-installs and auto-updates the exe from GitHub Releases, creates desktop shortcuts for both the app and the web UI, and exposes a desktop_launch tool to start it from the conversation.

Claude-style drag-and-drop/paperclip file upload with content sniffing, document-to-Markdown via Microsoft MarkItDown (built-in JS fallback), text inlining, and a read_document tool for agents.

Reference logged-in Chrome/Edge tabs and authorized local HTML with @ for reading, batched actions, screenshots and WebMCP; collect annotations in chat drafts with a draggable toolbar. Select running Windows apps with icons, prefer a matching installed MCP or approve installation, then fall back to Cua Driver. Supports text and key chords, paged control handles, minimized-window checks and optional action readback. Includes guided browser pairing and refreshed annotation controls; background input depends on the app.

Pauses the agent right before it sends the next model request of a tool round while keeping the composer usable: a bare Enter or the resume button lets it continue, and text typed meanwhile is sent along as a steering message; the pause never interrupts an in-flight API call and the model does not notice it.

Voice input plugin for DeepSeek Harness (dsh): a microphone button in the composer turns speech into a draft transcript, with a choice of speech-recognition backends, optional polish through dsh own LLM routes, and a native settings page.

Synchronizes DSH profiles and local plugin source archives through WebDAV and S3-compatible storage, with encrypted snapshots, conflict-aware recovery, and explicit self-updates.

DeepSeek Harness Cordis tools for All in Flash, powered by All in Luna

Unified MCP, skill and expert management board for DeepSeek Harness: connection status, tool schemas, enable/disable and reconnect for MCP servers across global and per-project scopes, enable/disable of SKILL.md skills, attach/switch/detach WorkBuddy expert personas in any conversation at any time (429 experts) with no mid-session preset lock, a tool tailor that scans local session logs for per-tool call counts, error rates and context-token cost with one-click disabling, and three install markets (SkillHub 190k+ skills, WorkBuddy connectors, 47 verified built-in services).