Plugins

Adds a "Delete session" item to the sidebar session menu that moves the session transcript directory to the Windows Recycle Bin, or to a plugin trash folder under ~/.dsh/plugin-data when that is unavailable (restorable either way), and deletes its spill files and projection cache in one request, detaching a live session first and cascading to its subagent sessions. A setting switches the confirmation step off so deleting takes one click; duplicate delete items registered by other plugins are collapsed so only one remains.

Runs sessions on your own Claude Code and Codex CLIs from the model picker, and lets a DSH agent hand background tasks to them and follow up. Permission prompts cannot be answered from DSH: Claude Code runs in acceptEdits mode by default, Codex in its workspace-write sandbox, and ACP permission requests are approved automatically by default.

Runs a four-task, 150-point model intelligence benchmark inside DeepSeek Harness (Qin 60, Pompeii 45, Pelican 22.5, Candy 22.5): each selected model × reasoning-effort pair executes every task in its own isolated subagent session, Qin and Pelican are graded by the selected judge model against rubrics, Pompeii and Candy programmatically, and batches keep scores, artifacts and cost. Adds a 0–1000 real-world performance score per model × effort computed from local session logs across seven dimensions (output speed, first-token latency, reasoning speed, tool cadence, tool latency, tool success rate, cache hit), with baseline deltas and controlled speed runs, plus a sidebar page with ranking, performance, history and settings views. Each batch spends real tokens of the candidate and judge models, and candidate subagents get file and bash tools inside an isolated run directory.

Rewrites a draft prompt against the current project: a composer button, slash command and agent tool ground the rewrite in the workspace stack, structure and conventions.

Gives the agent its own context usage through a context_usage tool plus one reading line per turn, with an editable template.

Shows the agent's live state in an always-on-top strip on the desktop — one ball when idle, a pill with the running tool name and elapsed time while working, expandable to the last message.

A draggable task progress panel with local estimates, independent teammate progress, optional AI observations, and probe cost diagnostics. Release candidate for the Web profile only; needs Node 24.9 or later and Harness >=0.2.2-rc.1, tested on 0.2.2-rc.2. AI observations are off by default and, once enabled, spend tokens of the session's provider. It uses the same package name and mount id (dsh-task-progress) as chen8923/dsh-task-progress, so install only one of them.

A reading room kept for the agent: through reading_* tools it reads a book from ~/.dsh/reading one chapter at a time, writes notes beside chapters and keeps its own writing there, and a panel shows its reading progress. It adds lines to the system prompt telling the agent the room is its own and optional. Not a to-do list.

Creates DSH workspaces under a configurable base directory, links several external folders into one workspace, and relocates a workspace without losing its conversations. Its managed_workspace tool lets the agent create, adopt, link, move and unregister workspaces (moving relocates the whole folder and leaves a link at the old path when it has conversations), and its HTTP routes (/dsh-managed-workspaces/api/...) have no DSH access check beyond a marker header.

Delivers one required user-role runtime context that makes the agent reason in Simplified Chinese and keep a whale-loli in-thinking persona (calls itself 本鱼, lazy-arrogant tone) in both reasoning and the final reply, while proper nouns, identifiers, paths, commands, URLs and error text stay verbatim; persona, extra proper nouns and context order are configurable. Enabled by default for every agent in the profile, subagents included.

Add one global prompt or persona that is injected into every new DeepSeek Harness conversation.

Native Google Antigravity agents loader and subagent orchestration engine with model selection and Web UI settings integration. Loads agent definitions from ~/.agents/agents and the workspace's .agents/agents; the selected main agent's instructions go into the system prompt, and a matching agent's instructions are prepended to DSH's own subagent and spawn_teammate calls. Settings are written to .agents/agents/settings.json in the workspace.

Agent gateway manager: picks a working gateway among proxies and direct, verifies the model channel, and switches online; DSH plugin plus standalone CLI.

Unofficial DSH bridge for Xiaomi Miloco: an isolated Agent preset for Miloco tools, native tools/skills, Web settings, perception-priority filtering, and Miloco -> DSH push.

Anthony Fu's agent workflow for DeepSeek Harness: issue triage, digest, task execution, PR creation and simplification

Read images through a dedicated vision subagent whose model is configured independently of the main agent. · 通过独立的识图子智能体识别图片，子智能体模型可单独配置。

Daily auto check-in for MiniMax Code, WPS Lingxi, Baidu DuMate, AStudio, Qoder, Trae, Doubao, Huawei CodeArts and Tencent WorkBuddy: reads each app's own local credentials so nothing has to be pasted, supports multiple accounts per service, and lets you add any other site yourself by importing a JSON definition the agent wrote for it. Streak tracking, per-service schedules, collapsible sidebar cards. The Doubao service can also read its login cookies from the local Chrome, Edge, Brave or Doubao app cookie store (Doubao hosts only), and a custom service sends requests to whatever http(s) URL its JSON names.

Event-driven wait for DeepSeek Harness: returns the moment the target completes, fails early, or stalls — instead of burning a full timeout and querying again. Also blocks sleep-polling and over-long waits.

Context-budgeted local memory for DeepSeek Harness: fixed ~600-token cache-friendly injection (session/daily snapshots), three-stage sleep consolidation, spaced repetition, temporal validity, subagent-driven L0 mental-model rewrite, typed memories (fact/s

Declare local executables, scripts, bash and PowerShell commands as Agent tools with exact argv mapping, hot reload and a settings page.

Nexi 灵犀：超级 Agent + 多 Agent 小组协作 + MCP 网关。一个会自己干活的 Agent，能把本机 CLI、ACP、OpenAI 兼容、桌面 App、自研桥接的多类 Agent 编成小组，并用一条命令把同一套能力接进 Claude/Cursor/Codex 等宿主。

Let your agent understand your players. Glitchless for DeepSeek Harness: game observability tools over MCP (ranked findings, frame times, errors, where players quit, each game's own dashboard) and skills that add Glitchless to a game in Unity, Godot, Unre

allow/ask/deny permission gates for DeepSeek Harness tools: path/params/agents surfaces, argv decomposition, network rules, redaction, grants, dry-run, fail-closed config

A session-aware code pane for the DeepSeek Harness right sidebar: browse, edit and run workspace files, see what the Agent changed, and step through every recorded edit. 代码面板 for DeepSeek Harness。