Plugins

DSH ↔ WeChat bridge: inbound dispatch (WeChat → DSH agent, persistent per-cwd session pool) + outbound push (task progress → WeChat via hermes send, zero-LLM). 微信 ↔ DSH 双向桥：入站调度 + 出站推送，常驻会话复用。

Inbound A2A (Agent2Agent) protocol server for DeepSeek Harness: publish an Agent Card and let any A2A-compliant peer submit tasks to a harness agent

dashi-taskboard 的 DSH 双面插件（独立版）：host 半在宿主进程拉起 SQLite server（数据目录 ~/.dsh/storages/dsh-taskboard），client 半注入侧边栏入口并把完整看板（看板/列表/Gantt/工作流/仪表盘/AI 对话）挂进 GUI 中间列。

Browser Use bridge for DeepSeek Harness: let your dsh agent run real web tasks (open pages, click, type, fill forms, extract data) through the Browser Use Cloud API.

DSH tool plugin: delegate a self-contained task to your local Claude Code subscription (via the official Claude Agent SDK).

A configurable desktop pet overlay for DeepSeek Harness with live task progress.

DSH Web taskboard plugin with workspace-scoped Agent sessions, staged human review, delivery confirmation, and durable archival

DSH plugin: project/task dual-mode workspaces. Tasks skip the workspace picker — every task conversation gets a fresh scratch directory under D:\dsh_working\<name>-<timestamp> (Codex-style), while projects keep the existing workspace flow. Ships an idempo

DSH tool plugin: generate images through ToAPIs async GPT-Image-2 API (submit task, poll, download).

Two-way TickTick/Dida365 (滴答清单) task sync via the official OAuth Open API: pull tasks across lists, create/update/complete/delete tasks, and one-shot title-deduped sync, with a web settings panel.

Native Windows system notification (Notification Center toast) when a task completes: goal completion or agent turn end.

Email reminders from your DSH agent when you're away: fires at goal completion, blockage, before asking the user a decision, and at long-task milestones via a bundled zero-dependency Node SMTP sender.

Short Web Audio cues when a session needs your attention (approval, question, or plan review) and when a task finishes. No popups or OS notifications.

Sound + popup when a task completes: a Web Audio chime, an in-page toast, and a system notification while the page is in the background (browser-only, cross-platform).

Desktop notifications and tones when a task finishes, waits for approval, or errors — native on macOS, Windows and Linux with zero runtime dependencies.

Adversarial code review — parallel finders inspect a diff through separate lenses (correctness, lifecycle, contract, security), then each finding goes to independent verifiers tasked with refuting it, and one refutation drops it.

Event-sourced cross-session work ledger: register, track and settle tasks with full audit history, kanban export, and a board digest injected into every model turn.

Declarative task-DAG orchestration for DSH: validates dependency graphs, runs topological task layers in parallel through workflow-backed subagents, and propagates failures deterministically.

Human-in-the-loop tools (`human_task` / `human_task_ready_check`): pause the agent until the user performs a real-world action or returns an observation (GUI operation, game testing, visual verification, hardware checks), gated by session consent and an AFK presence check, returning a structured JSON result via a web task dialog.

Personal toolbox for the dsh web profile with per-feature toggles: task-done desktop notifications, one-click restart, session deletion, plugin enable/disable, update checks, and UI utilities.

Scans a project's declared build commands — Makefile targets, package.json scripts, just recipes, deno tasks — and exposes them as ready-to-run agent tools.

Project memory for coding agents on DeepSeek Harness: similar past edits surface as precedents, out-of-scope edits are blocked, and cognition persists across sessions (47 upstream-verified fixes, gold-free 20-task run).

Cross-session event-cue memory for DeepSeek Harness: auto-extracts task/conversation cues and user language habits from every turn into a persistent global cue bank, then re-injects remembered context on topic-switch detection (keyword or vector matching).

Live per-task token cost widget: peak/valley pricing windows, projected post-hike costs, sortable per-task cost table, multi-currency (COP/USD/CNY + ~40 more) with FX rates, and one-click streamed LLM explain of expensive tasks.