Plugins

A `describe_image` vision tool for text-only models: images (local path, URL, attachment) go to a configurable OpenAI-compatible vision endpoint and only the returned text enters the session.

Native skin marketplace and lifecycle manager that discovers community skins, displays previews and compatibility status, and provides verified one-click or manual installation paths.

Multi-machine remote workspace: manage many SSH hosts, pick a local or remote workspace in the native Add-workspace flow (system folder chooser / local path / remote dir browse), mirror a remote workspace to a real local folder, and operate it with rw_* tools. The picker is a centered modal that opens on the local tab and auto-fills `/` for remote paths, live-completing each directory level.

Red-team engagement mode: send one target organization name and a planner session runs a five-role execution team (recon, asset triage, vulnerability discovery, exploitation, internal pivot) under a configurable concurrency limit; preflight checks skills/resources and asks once for missing keys or VPS; scoring follows the 8-category / 25-point merged intrusion rules with server-side caps, highest-privilege-wins and per-service dedup; findings stay in a local SQLite fact base with discovery times. The persistent right-side console has five sections / ten pages (situation, assets, findings, report, resources), follows the selected session, and includes asset mapping, agent roster, tunnels, findings, attack chain, score targets, reports, a category-organised POC/EXP knowledge base and a skill library that explains every unusable skill. Reports now download with the target's name, include the same full actions/requests/credentials/tunnels/attack-file paths as the page, and avoid false reproduction gaps. Ships 16 bundled recon/assessment skills including DeepSeek Search/Fetch public-intel, 56 redteam_* tools and one-command self-update; attack-chain skills remain a separate toolkit Release attachment.

Claude Code-style declarative permission rules: ordered allow/deny/ask YAML rules matching tool names, arguments, workspace paths, and agent identity on the tools/pre-execute waterfall, with full session-log audit, dry-run mode, and hot reload.

Read-only session work-path observer: fold turns into stage, run, duration and tok/s summaries, then expand parallel branches, tool results and vendor-visible reasoning.

Causal memory for dsh: a zero-dependency MCP stdio bridge that mounts the 17 tools of the causal-memory server under clean names (no mcp__ prefix) — decision → outcome edges with typed relations (caused / enabled / prevented) in a local SQLite store that survives compaction — plus a pinned system-prompt section. Requires the causal-memory server binary on PATH (pip install causal-memory, or a prebuilt release).

Visualize the DSH process network path on Windows or WSL with layered DNS/TCP/TLS/HTTP probes, detect stale proxy configuration, and apply snapshot-guarded repairs.

Execution maze for DSH sessions: the agent's real path — main advance, failed or dead-end detours, backtracks, request retries and subagent branches — drawn on one wall-clock timeline, with three tracks under each lane sharing that axis (tool-call density, token pulse, context pressure) and a deterministic analysis panel below it (tool result matrix, failure-recovery chains, duration distribution); watch it grow live in a session tab, or upload up to 5 session logs for a same-axis compare.

Trace Compare & Live Maze: draw the agent's real exploration — main path, failed or dead-end detours, backtracks, and subagent branches — on one wall-clock timeline; upload 1–2 session logs for a same-axis compare, or watch it grow live in a session tab.

TapTap Maker integration for DeepSeek Harness: bundled MCP tools, Maker workflow and feature skills, plus a packaged CLI path for project initialization and diagnostics.

Puts the Markdown documents, source files, images and video that already exist anywhere in the session workspace into the DSH sidebar, ranked by relevance to the current conversation. The default lexical mode matches your recent messages against title, summary and body locally with IDF weighting (BM25) — no model calls, no network, nothing leaves the machine. The list comes from scanning the workspace rather than from a recent-files list, so restarting DSH or opening a new session does not empty it. Results are organised into a three-tier reading order (primary / supporting / related) with a verifiable reason on every row — which field matched, which document cites this one — and one click opens an in-place preview: Markdown as rendered text, relative-path images actually drawn, source code with the host's own syntax highlighting, video streamed over HTTP Range. The same Context Pack is exposed to the agent as a knit_docs tool. Later versions added read evidence and lifecycle tracking (attribution is frozen at the moment of reading), pin / exclude (exclusion applies before retrieval, and is undoable), relation evidence (references / imports / tests, with line numbers) and an incremental index reused across restarts. Since v0.23 there is also an optional hybrid mode: it is off by default and only runs once you configure an Ollama or OpenAI-compatible embedding provider yourself. When it runs, what leaves the machine is your query text plus each candidate file's title and summary — a Markdown summary is an excerpt of the body's first paragraph (at most 60 characters), a code file's summary is its workspace-relative directory, and an image or video contributes its file name; full bodies and absolute paths are not sent, and a loopback provider keeps everything on this machine. Since v0.24 it also associates uncommitted changes: a read-only git status (fixed argv, no shell, whitelisted child-process environment) combined with the same reference / import / test / mention evidence shows which files a change is likely to drag along, and it is off by default. The benefit of semantic retrieval still awaits a real-provider run, so no recall improvement is claimed.

Claude Science-style research workbench: ReAct research-loop engine (research_* tools), versioned artifacts with provenance (artifact_* tools), and 10 science skills for genomics/pathogens/bioinformatics.

DSH↔Godot engine runtime bridge: launch and drive a running Godot 4.x game through its in-game TCP interaction server — 16 tools (runtime control via ~130 game commands, scene/UI inspection, GDScript eval, input simulation, screenshots, headless static ops, project editing, export, script validation, engine-path setup), replaces godot-mcp.

Plays a video while DSH opens, then dissolves over 0.3 seconds once it has played out. Ships with a default video so it works on install, and the path can be changed or cleared from the plugin settings page.

Live2D desk pet overlay for the DSH web GUI: mirrors agent state (thinking, idle, error, done, awaiting approval) with motions and speech bubbles, per-part touch reactions, drag-to-dock, switchable personas, and custom models via URL or local path.

Paper reading workbench: library + PDF reader + tutor-style companion. Answers cite clickable page numbers that jump back and flash the exact passage; select-to-ask without leaving the paper; bundled strict-tutor preset (reading plan, section coaching, graded quizzes tracked across sessions). With MinerU configured (self-hosted, no signup / mineru.net cloud token): scanned PDFs become readable, formulas become click-citable (one click sends the LaTeX source into the chat) and searchable (QK^T hits directly), plus a per-page formula panel; optional embeddings endpoint adds semantic recall for plain-language questions. Rename/delete topics and papers — deletes go to a recoverable .trash with a batch-clear flow for non-empty topics; reader with scroll/single-page modes, rightbar-fullscreen safe, one-click path copy and translated-PDF export; in-place original ↔ Chinese toggle (babeldoc, zero preinstall). Transcription and search run locally by default — documents never leave your machine; only when cloud MinerU, an embeddings endpoint, or a translation endpoint is enabled does the corresponding text go to the service you configure.

Injects a fixed system-reminder at the first step of every turn asking the model to reason and answer in Simplified Chinese, while code, commands, file paths, and identifiers stay in English.

Cross-platform drag & drop with original path insertion.

Cross-platform file drag-and-drop with raw path insertion, no file copying.

Memory profile composition with Chinese session search, hybrid recall, persistent workspace facts, user-pinned rules and reusable native skills. Install the standalone versioned Release bundle with one dsh plugin add command. Requires Node.js 24 and pnpm on PATH.

JetBrains-style Git tool window for the sidebar: branch tree with ahead/behind badges, commit graph with text/author/date/path filters, per-file line diff, changes list and commit box, plus branch, tag and stash actions, merge, rebase, cherry-pick, reset and fetch/pull/push.

Clickable file paths in DSH replies: Codex-style inline open, reveal in file manager, and a mentioned-files chip list at the turn tail.