Plugins

Composer and sidebar helpers in one plugin, each switchable in Settings: a git Changes panel, Turn rollback, Resume after an AIProxy outage, Project memory (notes per project that later chats read, off by default), a Task board (kanban board of chats by their live state), Assistant (one lasting chat named Damon that sets reminders with dsh scheduled tasks and reads your other chats; it changes a chat only after an approval) with Assistant callouts, Assistant always free (long steps go to background subagents) and Assistant news from chats (off by default), Remote computers (another Mac's dsh over Tailscale, off by default), 11 extra UI languages and Automatic updates.

Agent presets whose leader does no work itself: every non-management tool call is denied at the harness boundary so the task has to go to a subagent, after a one-time lookup of the available subagent models and a per-task model choice; a second preset deliberately picks the model least suited to the task.

Send dsh turn traces, LLM generations, tool spans, and subagent activity to Langfuse.

Splits New Session in the DSH web sidebar into two modes: a 日常聊天 row opens a session on the `daily` agent preset the plugin installs into the harness home (persona, web search and fetch, ask-user and compaction — no files, shell, plan mode or subagents), inside its own daily-chat workspace whose sidebar region lists only those chats, while 开始工作 keeps the shipped coding agent; the built-in New Session button is hidden and every other New Session entry point follows the active mode, which is remembered in localStorage.

Kanban task board for DeepSeek Harness sessions: the main agent dispatches tasks to one-shot worker/verifier subagents, with dependency scheduling, file-level exclusion locks, arbitration hold/freeze, milestone progress reporting, and per-task token usage stats.

Smart subagent routing and evolution: bind roles to any provider/model registered in DSH, including models supplied by Codex and Claude Code provider plugins, with workspace learning.

Named subagent roster: manage role entries (model, persona, tool filter, depth, background mode) from a hot-reloaded settings page; the model picks one with list_subagents and dispatches by id with delegate — foreground, background, or continuable.

A thinking level stored per provider and model, so reasoning effort outlives the session it was set in and also reaches subagents, which dsh never asks. The in-session picker still wins, and a model with no row keeps its provider default.

Per provider/model generation concurrency, with a settings card: a cap on how many sessions may stream at once, enforced at the stream and again at subagent spawn, where a fan-out past the cap waits in a FIFO queue instead of failing.

Load Claude Code assets into DSH at runtime — skills, slash commands, rules, permission rules (deny/ask/allow), subagents and hooks from .claude/, plus Claude Code plugins (plugin.json / marketplace / skills / agents / MCP), via an in-memory IR that keeps .claude as the source of truth and writes back only when you opt in. Also surfaces Claude Code auto-memory in the system prompt, read-only by default, with a write-back switch in the /cc-memory panel. Eight packages: one bundle that installs the whole set, plus the loader and six adapters (skills, permissions, agents, hooks, mcp, memory).

Engineering workflow agent preset for dsh: five gated phases (requirements clarification, plan approval, TDD, parallel subagents, verified finishing) with six workflow skills adapted from obra/superpowers.

Cross-session archive: overview workbench, folded timeline, subagent lineage, full-text search, rule+LLM summaries, ZIP export, live tracking. Native session-page embeds: turn rail, docked right panel (live mode, pinned to original session on switch), rename (official-synced), conclusion digest, health monitoring + persistent header chip (idle / active / tool-wait / stalled). dsh v0.1.2+: fuses with the official TurnNavigatorRail — vector-icon info chips (duration, tool calls, errors, thinking) plus lazy tool distribution inside the official preview card, right-click opens the workbench; the classic piano-key strip is on by default and delegates unloaded-turn jumps to the official auto-load engine; the official rail can be hidden entirely. Settings dialog section with 10 tunable behavior params.

Office workspace & session dashboard: a floating 6-column sprite panel visualizing workspaces, sessions, token usage and subagents, plus Agent Mail and an office log tab. Feishu/Lark views (message feed, meeting schedules, transcripts) additionally require the `@larksuite/cli` tool installed and authenticated separately.

Telegram bridge for DeepSeek Harness: forwards agent questions (with inline buttons), tool approval requests, turn summaries and subagent messages to a Telegram bot, and takes the answer from Telegram or the DSH Web UI, whichever comes first. Ships a loopback control panel; host-only, no client UI.

Adjust Agent Team's member count, task count, message size and teardown timeout.

Keeps the PTC run_code deadline from firing while DSH waits for your answer, plan review or approval, so a question can stay open longer than the shipped 120 s budget.

Find keywords in the current conversation or across every session, with highlight and jump.

Three tools that let any agent, including a team member, list, steer and stop the subagents it started.

Deep-reading workflow for books: import EPUB/Markdown/TXT/PDF into chapters, read one chapter per subagent so the source text never enters the main session, keep \[n]-anchored chapter memories for cross-session continuation, and export notes or an EPUB.

Windows Computer Use for DeepSeek Harness with 21 tools and verified targeting: controls are addressed through their UI Automation identity and patterns, and numbered marks are checked against the live window before input, so a stale target is refused instead of clicked. A known sequence runs in one computer_batch call, one model round trip; incremental observations, compact receipts (normally no more than about 6 KiB extra), bounded handoff waits (HANDOFF_WAIT_TIMEOUT/HANDOFF_HELD_TIMEOUT as continuations), the desktop-block cycle boundary and a deterministic 100 s owner-scoped leave backstop keep context small and prevent a stuck handoff from waiting out the tool deadline. Approval modes settle an unanswerable ask honestly: in a danger-full-access session with no approval card the action runs under that grant, every other sandbox — and every delegated subagent session — refuses and names the unavailable channel instead of reporting a user rejection, and gate diagnostics land in policy.jsonl. Human handoff, a model-visible stop brake and prebuilt native workers.

Make your dsh ready for serious coding. (Tools x Schemas)^REPL. `skill://`, `ctx://`, `agent://`, `dvc://`, `dsh://`, IPython REPL, Context as Variables, cross compactions recallable, full context revive. `hash-edit`, `dvc://browser`, subagent as function, workflow as function.

Make your dsh ready for serious coding. (Tools x Schemas)^REPL. `skill://`, `ctx://`, `agent://`, `dvc://`, `dsh://`, IPython REPL, Context as Variables, cross compactions recallable, full context revive. `hash-edit`, `dvc://browser`, subagent as function, workflow as function.

A customizable widget rail for the right side of DeepSeek Harness: 61 cards covering session stats, tokens and cache, context, background jobs, subagents, machine metrics, OpenCode Go and Command Code quota, session cost, peak pricing and the GitHub review queue. Cards install from an in-rail market into 1-4 column grids and 2x2 or 2x4 tiles, with per-instance settings, a magnification wave under the pointer and a one-click hide. Works on DSH 0.1.x and 0.2.x, web UI and desktop app.

An independent approval subagent judges every sandbox escalation, with a configurable model and an audit log.