플러그인

Workspace-free chats that promote real files into their own project workspace - a maintained fork of dsh-scratchpad, kept working on current DSH.

Sidebar layout fix for the DSH Web UI: caps and scrolls the plugin-action area, shrinks the inline pairing QR code and tightens the brand row, so the workspace and session list stays clickable when footer plugins grow tall.

Chinese summaries for the dsh-skill-mcp-panel skill cards: each expanded card gets a short Chinese overview, a category tag and a few bullet points, while the English description the panel already renders is folded behind a "原文" toggle. Skills missing from the bundled map are translated on demand by the host with your current default model and cached on disk, keyed by a hash of the original text, so a new skill needs no manual entry. Skills whose own description is already Chinese are skipped.

Selection comments and quoted conversations (dsh划词评论 / dsh-huaci-pinglun), based on the MIT-licensed BangBang-03/dsh-client-ui-quote 0.11.1 with attached textarea measurement, border-aware height and long-text scrolling.

Adds a session status pill to the composer dock that expands into a panel of turn, token, performance and context readings.

Right-to-left reading mode for the conversation answer area, with a button to switch back to left-to-right.

Adds three visual pages to DSH Settings: MCP servers (managed by rewriting the profile patch managed block from a JSON source of truth, with YAML validation and timestamped backups), skills (browse, create, edit and delete across DSH global, shared .agents and Claude / Codex directories) and the global AGENTS.md persona (with one-click restore of the last three backups).

Executor/advisor workflow ported from pi-advisor-flow: the executor calls ask_advisor for a second opinion at key decisions, after repeated failures and before completion; equivalent repeated tool calls are intercepted before execution for an advisor review with a proceed/revise/blocked verdict, and hard mode force-reviews every turn before it closes. Per-call token and cost ledger with /advisor status readout, privacy tiers for repo context and file contents, six-shape secret redaction, and a web settings card; advisor failures never stall the main loop.

MCP manager for DeepSeek Harness Web: a Settings page beside Plugins that lists the MCP servers the running loader tree actually holds, each with its config summary, endpoint and live state, and disables or restores any of them through a paired-marker block in the profile's cordis.patch.yml. Env values, header values and `!!js` sources reach the page only as key names or dot-paths.

Claude-style composer header bar: shows working-tree diff counters, quick branch switcher, native diff drawer integration, and adaptive PR creation with Actions CI status.

Subagent model control: a Jev classifier labels each turn, and the subagent request itself is rewritten to the provider, model and reasoning effort that the label maps to, at the agent/request hook; a per-session inherit/auto/manual selector, one prompt per role injected ahead of the task, and offline replay of past verdicts.

Export DeepSeek Harness (dsh) sessions as readable Markdown. Pick individual messages from one session, or merge several into a single .md / split them into a ZIP.

Removes sandbox_permissions / justification arguments that cannot widen the call's current sandbox mode, so calls no longer fail with 'sandbox escalation to "workspace-write" is not strictly wider than this call's current "danger-full-access" mode'. GPT models tend to fill these optional fields on every call and pick workspace-write, which is narrower than danger-full-access. Runs in tools/pre-execute: compares the requested mode with the mode the session's sandbox policy resolves, and drops the fields when the request is the same or narrower (also drops a lone justification). Also understands the values GPT carries over from Codex's own shell tool: null, empty and use_default are treated as no request, and in a danger-full-access session any value (including require_escalated) is dropped because nothing is wider. Real widening requests, unknown modes and tools that do not advertise DSH's escalation schema are left to DSH core. Never widens permissions. Configurable tool list (default bash, pwsh, write, edit, run_code). Zero dependencies.

Fork of amlyczz/dsh-lark-link (MIT), maintained by Very12345. Per-conversation reasoning-effort controls, multi-app isolation, and streamed task-status cards.

Discord bridge loaded into the same Host as the desktop app or dsh web, with no separate process or profile — chat with sessions in any workspace from DMs or threads, attach to sessions open on the desktop, and answer tool approvals and model questions from whichever side responds first.

Pushes DSH session events to phone IM and lets you reply straight under that message, injecting the text back into the right session. Covers nine event kinds: task completion, execution error, execution blocked, manual abort, token limit reached, abnormal interruption, waiting for your answer, permission request, and waiting for plan confirmation. Channels are QQ single-chat bots and Feishu custom apps, both over WebSocket long connections — no public IP, domain or tunneling needed. Several bots can be configured at once, each subscribing to the sessions it cares about. Approvals and questions arrive with tappable buttons, and replies route home through three layers (button, quoted reply, explicit short-id prefix), falling back to the most recent intervention-needed notification and echoing which session it went to.

Bridges DeepSeek Harness sessions to Feishu/Lark, Telegram and DingTalk: each chat is bound to a resumable DSH session, replies stream back as cards, and a task-end card reports the model, token usage, step count and duration. Question and permission-approval cards are answerable from the chat, and a web settings pane configures the per-channel credentials.

Five task-state reminders for DeepSeek Harness (completion, approval, question, block, error) with four synthesized browser sound presets, system notifications that open the right session, and a terminal-BEL fallback when no browser is ready.

Source Control panel in the DSH Web left rail: search, switch and create branches with ahead/behind counts, a topo-order commit graph, and a click-to-expand preview of any commit showing its files (A/M/D), insertions and deletions, message, author and timestamp. An always-on footer dock above Settings carries the branch dropdown and recent commits and follows the open session; fetch and fast-forward pull are one button, and a git_worktree agent tool creates, lists and removes worktrees.

Agent skill and tool for querying multi-platform model account balances: DeepSeek, Doubao / Volcengine, Qwen / Alibaba Cloud, Hunyuan / Tencent Cloud, and custom HTTP endpoints. Read-only, with 60s cache.

Speaker button left of the model selector in the composer tool row — one click scrolls to the start of your newest question, marks it with a blinking caret and reads from there through the newest agent reply (dsh-tts, browser voice as fallback); press-and-hold the icon and then click anywhere on the page to pick another reading start position; press-and-drag-up opens a vertical mixer for in-page media volume and system output volume.

Gives DSH the full 33-tool browser surface (ctx.browser seam -> a self-hosted headless moli browser over CDP: one 127.0.0.1 daemon with Bearer-token auth, request-path whitelist and WebSocket rewriting, no root/GUI/Electron needed; single-engine moli serve with auto LD_LIBRARY_PATH/FONTCONFIG_FILE injection and a persistent profile dir for logins), plus a web panel — a real page you can click, scroll and type in, operation/console/network logs with one-click cleanup, a customizable window (border color, transparency, glass, fit-to-picture with no black bars, top-inset and self-centering) and an adjustable page size (1920x1080 by default) — plus a collapsed pill that can be dragged to any spot on the screen (free position), a live view that auto-resumes when you switch back to the web tab, and a 1920px default max edge for the live frame — and two bundled skills (browser, browser-runtime) served through the plugin built-in skill provider, so the skill centre lists them as system-level by default without writing into $DSH_HOME/skills.

RSS/Atom reader with grouped subscriptions and unread/total counts, FreshRSS account sync over the Google Reader API with read/star written back, local image proxy, optional BYOK AI summary/translate/Q&A, podcast audio playback and full-text fetch.

Mounts the hosted Reaudit MCP server (https://mcp.reaudit.io/mcp) through the in-box @deepseek-ai/dsh-mcp-client bridge with a REAUDIT_API_KEY bearer token: 238 mcp__reaudit__* tools for brand visibility in AI answers (ChatGPT, Perplexity, Gemini, Google AI Overviews), competitor and citation analysis, site audits, content generation and social scheduling.