Plugins

Plugin governance: new/updated plugins stage in a candidate area first, then hot-mount after startup — success promotes, failure auto-disables, repeated failures freeze, one-click safe mode, sidebar diagnostics panel (npm: `dsh-my-guardian`).

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: `dsh-my-notify`).

Task reliability: auto-retry on model timeouts/failures, auto-continue unfinished tasks, completion verification, thinking-repeat interruption, resume after sleep or restart, autopilot mode and a remote trigger endpoint (npm: `dsh-task-reliability`).

From chat to canvas to artwork — drive ComfyUI as a visual workflow IDE inside DSH. Embed ComfyUI (local or cloud) as a split-screen canvas in DeepSeek Harness Web: the agent sparks ideas, writes prompts and scripts right in the chat, applies them live to the canvas in front of you, and produces images, music, video, and 3D. From idea to finished output without ever leaving the conversation or switching front-ends: Canvas ops — compose and arrange pipelines, read/write workflows, edit nodes, wire links, run, tune parameters, and debug errors, all live and WYSIWYG on the exact canvas you are looking at; Production tasks — batch parameter sweeps (batch_run) and automatic output-image retrieval back into the chat (get_outputs), powering multi-modal creative and batch generation across images, music, video, and 3D; Environment upkeep — one-click launch of ComfyUI and one-click upgrade of the core plus every custom node (upgrade), keeping the stack healthy without interruption. This package is the DSH-side plugin, and it ships the ComfyUI-side bridge node too.

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.

DSH sidebar file-activity tab: records file reads/creates/modifies by the agent and the sidebar (including file ops hidden inside bash commands), per-session isolation, survives restarts, click a file for a floating preview (npm: `dsh-file-activity`).

Renders mermaid/mmd code blocks in the conversation as chart cards with preview/code toggle; the engine is inlined so it works fully offline with no CDN (npm: `dsh-mermaid-render`).

Thinking in Chinese: injects a system prompt so thoughts and replies stay in Chinese, expands the reasoning block by default with interactive collapse, and localizes UI labels (npm: `dsh-think-zh-expand`).

Fold completed turns into their final conclusions, split thinking from prose with a divider, and jump between turns from a rail dockable left or right with line or dot marks, an optional ring and hover summaries; it can hide the host right-edge TurnNavigator, load earlier history, and open a pre-filled diagnostics issue.

A whip for DeepSeek V4 Pro: it watches the reasoning fingerprint of every thinking block and pulls the model back when it slips from focused "We will / I will" mode into scattered "let me" mode.

Two-way WeChat (iLink) bridge: turn-end and approval-request push, in-chat approve/reject and message injection, persistent dedup and convergent long-reply chunking; channel layer extensible to other IMs.

Per-round reasoning_effort optimizer for dsh — auto-downgrades reasoning for simple tool calls and lifts back for heavy work, cutting thinking time between tool calls.

Visual round-table meetings for DeepSeek Harness — expert nodes on a ring with directed edges, an aggregation gateway, proxy thinking for black-box worker models, and adversarial review (red-team) of settled plans with per-viewpoint endorse/reject.

Bridge dsh session status to the cc-pet desktop pet bubble: thinking, tool calls, and completion in real time. Ships on npm as `dsh-pet-bridge`.

Guarded context compaction for DeepSeek Harness (dsh): the LLM proposes, deterministic guards dispose — eager per-atom shrink (extract/summary/false under verbatim guards) + lazy reference-graph eviction (0-LLM) + byte-exact recall from an append-only log.

Git worktree panel for DSH Web: replaces the workspace sidebar with project → main/linked worktree → session groups, creates and removes worktrees, switches the main branch, and migrates configurable project-local or global storage with dirty-tree and active-session safeguards.

Per-round thinking-level (reasoning_effort) control for DeepSeek Harness: pick Auto and it schedules low/high/max from the recent tool-call history, or fix a wire level (off/on/minimal/low/medium/high/xhigh/max) manually, with custom wire mapping, a model-aware capability guard, and context-window presets (64K–1M); its per-model capability card pairs with a separate plugin, [dsh-llm-openai-completions](https://github.com/drscrewdriver/dsh-llm-openai-completions) (installable from dsh-market), so custom OpenAI-compatible gateways (vLLM / LM Studio / self-hosted) can try reasoning — the card writes llm-pi-ai capabilities, the adapter drives the wire (compat.thinkingFormat), and this plugin auto-maintains the takeover list.

Low-cost video understanding tool: a video_understand tool turns a Bilibili link / BV / local video into an AVIS info layer (ASR + scene structure + object tracks + YOLO labels) and returns summary + Q&A. Question-driven dynamic routing across layers (L0 ASR / L1 object tracks / L2 key-frame VLM), semantic-layer reuse for repeat questions, per-question budget cap. Python engine: core layer needs faster-whisper / opencv / yt-dlp (~200-300MB); the optional semantic layer adds ~2GB of torch / transformers / ultralytics. A bundled doctor --fix sets up the venv and installs both.

Settings card for the DSH LLM auto-retry engine: tune retry count, backoff and jitter live from Settings -> General, with configurable retryableCodes chips - defaults add INVALID_REQUEST so OpenAI-style HTTP 400 reasoning_text errors in thinking mode get retried automatically.

Fetches any http/https page and returns title, description, main text, heading outline and links, with stats (char/word counts, reading time, language) and top keywords; can download content images to the session workspace for vision analysis and dispatch them to registered analyzers via the scrape.imageAnalyzer service.

Global Chinese mode: a compact pill switch beside the composer that injects a language requirement into any preset's system prompt (complete personas excepted); reply, thinking and tool output each toggle between Chinese and English, and anchored presets are supported.

Auto-migrate Codex desktop-pet skins into DSH: an animated corner pet that mirrors agent state (thinking, tool use, awaiting approval, failed, done), with per-session frosted-glass dialogs and a full settings panel.

Solarized light theme, compact layout, think/tool-chain collapse capsules, and balance, session cost, and usage dashboards for the DSH web UI.

Streaming-transcript fold for the DSH web UI: thinking and tool rows fold away while the newest thinking window opens itself; a finished turn scrolls back to your question, which also stays pinned at the top while you read; blue sparks fly from the thinking window edge and the answer writing head.