Plugins

A ✨ button left of the send button: reads the draft and recent turns, asks the configured model for three suggestions each pairing an issue with a patch, lets you tick them individually with a live preview of the merged prompt, and writes the result back to the composer without sending.

Standing instructions edited in the DSH settings UI and injected into the system prompt of every subsequent turn.

Peer communication between conversations in one DeepSeek Harness process: list the conversations the workspace shows, link one under a nickname, message it without waiting for its turn to end, read its progress without waking it, and declare rules for your own tool calls.

A DSH plugin that turns a completed conversation into a reusable, validated DSH Skill bundle.

Turns DSH permissions into 9 selectable presets (3 sandbox modes x 4 approval tiers), with rule plus LLM risk classification across four risk levels, password-gated approval for high-risk operations, and a JSONL audit log.

Turn-level workspace undo for Git worktrees: every agent turn gets a private snapshot, /undo previews a red/green diff card that is confirmed in place, files changed after the turn are refused with a conflict diff unless --force or --skip-conflicts is passed, and HEAD, index, stash and commit history of the user repository are never written.

Minimal DSH Web transcript toggle: hide thinking, tool-call, and turn-process rows from the session header while preserving assistant answer text.

DeepSeek Harness plugin: turn lines of narration into a finished vertical reel. Free neural voice-over, burned-in captions, no API keys.

Live subagent progress dock for DeepSeek Harness: folds each child session's events into a subagentProgress projection (turn/step, last tool, latest assistant preview, notify_user updates) and renders a frosted-glass status dock above the parent conversat

Persistent plan-mode toggle button in the DSH Web composer: a two-state control (Plan / Plan ×) that turns plan mode on and off, replacing the shipped show-only-when-active badge.

Pressing Enter with an empty composer steers the first queued message into the running turn while the agent is busy.

Standalone dsh plugin (maintained by dsh-external): review agent-produced file changes (line-level red/green diff + undo) as a chat turn-tail row and a coding-sidebar tab. Lineage: originally a port of left0ver/dsh-file-review; previously maintained in-tr

A per-workspace HindSight switch for the DeepSeek Harness composer: turn memory off for one workspace without uninstalling it anywhere else.

Photo pet plugin for the DeepSeek Harness (DSH) Web GUI: turn any uploaded photo into a floating pet that can be dragged, dressed up and chatted with. 照片宠物插件:把上传的照片变成 DSH Web GUI 里的浮动宠物,可拖拽、可换装、会说话

Auto-send every DSH turn result to WeChat: listens for task completion, extracts the final assistant message and pushes it raw via the WeChat iLink bot (real delivery check, never fake success). Requires the wechat-acp daemon, a separately installed process you log into WeChat with.

Turn-end notification sound for DSH: built-in internet meme voices (ni gan ma ai yo, ji ni tai mei, shen ying ge, etc.), Settings panel with preview/switch/random, custom audio folder support.

Rolling, integrity-verified backups of each session at turn boundaries: verifies the log still loads before promoting a snapshot, never overwrites the last good one when a turn ends corrupt, and restores in one command even when dsh will not boot.

Explains which layer killed a failed turn: times every model request (first byte, longest silence, total) and combines that with the error to name the culprit (dsh idle watchdog vs Node undici timers vs SDK timer vs server crash, context overflow, gate queue, tool timeouts, failed compactions) and the exact setting to change; verdicts in the DSH terminal plus a /why command, subagents included.

Context-window attribution for DSH Web: a Mini Utility Dock panel that ranks the user messages, assistant turns, reasoning blocks and tool results of a session by the tokens each one occupies, so it is clear which entries hold the window. It also shows the share by entry type, the heaviest entries across the live sessions of the host, and, where the host provides them, provider usage and the estimated system/tools/messages composition. The panel is read-only: it does not estimate cost, modify a session, or trigger compaction.

Tools guard that moves agent-issued `rm` targets to the macOS Trash instead of deleting, with a shell-aware lexer covering compound and disguised commands; a switch in Settings → General turns it off.

Auto-adjust model reasoning effort per task (off/low/high/max): task classification, \[effort=...] markers, in-turn error escalation, DeepSeek peak-pricing-period capping with countdown notices, and manual-selection priority.

Auto-approval agent for DSH permission requests: rule layer (denyAlways/allowlist) plus Hermes Pro semantic adjudication with quick-fail retries, fail-closed to human when Hermes is unavailable, rejection feedback returned to the requesting session, and approve tasks enqueued into the task queue with a shared busy mutex.

ServerChan3 (Server酱) push notifications when a turn finishes — title, model, project directory, git branch, status and a reply excerpt, with subagent filtering and fire-and-forget delivery.

Task-complete and question notification sounds for the DeepSeek Harness Web UI: a soft chime when the agent finishes a turn, a ding-dong when it asks you something (questions, plan reviews, approvals). Synthesized Web Audio, zero assets, mute toggle and volume slider in the session header and settings.