Plugins

Source-backed logic reviews with diagrams, code diffs, and resumable human feedback. The first activation downloads the official LogicMargin release (SHA-256 verified); on macOS and Linux this needs curl, unzip and shasum or sha256sum.

Runs the agent through GEML workflows: work items move through flow steps, practices are recorded at the step they belong to, and a board tab in the right sidebar shows where each item stands. Also mounts the GEML MCP server and skills; when the project has a fixed workflow, a supervisor gates tool calls by its current state. Needs Node.js 22+; the GEML MCP server is fetched with npx (@geml/geml) when a session starts.

Task-pack builder for DeepSeek Harness: compress large files into one brief (default budget 8000 tokens, estimated) that a subagent can read alone.

A configuration-only DSH bundle that connects the MCP server shipped with Scrapling to DeepSeek Harness over stdio, exposing its HTTP fetch and stealth-browser scraping tools for page extraction, browser actions, screenshots and network inspection. Requires Python 3.10+ with scrapling\[ai] (scrapling-mcp on PATH) and a Chromium-compatible browser (scrapling install).

Zero-config per-Session freshness for codebase-memory: on Session start and on every working-directory change it refreshes the index of the Session's own repository, so each project in each Session answers from its current graph while indexing stays in the shared daemon.

Connects the Tabby serial MCP bridge and registers a tabby-serial skill: the agent can list, open, read and write the serial ports already open in Tabby, and the same terminal keeps accepting your keystrokes while it works. Requires the tabby-serial-mcp plugin inside Tabby.

Connects Penpot to DSH across two independent channels with 9 tools: 5 MCP tools (execute_code, high_level_overview, api_info, export_shape, import_image), 2 browser-free REST tools (penpot_read and penpot_write to create files, pages and shapes or edit attributes without an open tab), plus penpot_status and penpot_control; supports official SaaS, local and self-hosted endpoints with auto-detection and config migration from DSH, Codex, Claude, Cursor, Gemini and .mcp.json, an isolated Chromium launcher, and a five-tab control panel.

Drives Windows apps through accessibility controls instead of pixels, with a takeover overlay that frames the window being driven, marks each action with a ghost cursor, and stops on Esc; a browser lane adds sign-in profiles, saved state, and attach-to-my-Chrome. Windows only, and DSH must be run as Administrator for the desktop lane.

Sends deferred MCP tool definitions as an index of names, parameter signatures and one-line summaries grouped by server, with tool_details returning a full schema on request and call_tool running the real tool. Reading a schema or running a tool does not change the model-visible tool list, so the prompt cache is not invalidated.

Microsoft To Do panel: browse lists, add, edit, complete and reopen tasks, and open notes, due date, reminder, recurrence, steps and attachments. Lists can be collected into local groups, and a batch of tasks can be picked into a My Day shortlist. Signs in with the Microsoft device-code flow; the token cache belongs to the upstream microsoft-todo-mcp server, pinned to 0.2.0 and started with npx (so Node.js and npm access are needed) with only an allow-listed environment, while groups and My Day live locally. Today, Overdue, My Day and new due dates and reminders follow the device time zone (override with DSH_TODO_TZONE). Chinese and English UI.

Fork of SheepofSheep/dsh-doyourweb that puts the browser on another machine: the MV3 extension dials out to this plugin over TLS, so DSH reads and acts on a remote Chrome or Edge rather than a local one. It also generates ready-paired extension packages for the current configuration, adds tab open and close tools, writes annotation detail to a file instead of pasting it into the chat, and shows a checklist naming which part of a remote setup is misconfigured. Interfaces with no DOM at all are reachable too: canvas games, canvas maps and hand-drawn widgets take coordinate clicks, drags and wheel input that glide, pause and hold like a person's, and the viewport is checked before every action so a stale coordinate is refused rather than clicked blindly. It reuses the same browser-tabs plugin id as the upstream, so do not install both in the same profile. Requires Chrome or Edge 125+ and an unpacked MV3 extension loaded manually on the browser machine; the plugin does not install the extension. With tls: true, trust the generated ca.crt as a root certificate there. The extension gives the model access to the tabs and signed-in sessions of the browser it is paired with. The bridge listens on 127.0.0.1 by default; a non-loopback bind needs remote: true, and the extension pairs with a token confirmed by the user.

DSH slash command /rea: hand a reverse-engineering request to the reverse-engineer-anything skill and the REA MCP tools, with a titled and glyphed / menu row

Lightweight local visual grounding engine exposed as agent tools

Aggregated MCP market for DeepSeek Harness: 26 curated presets plus the official MCP Registry and user-added JSON sources, with search, source filters, connectivity tests and one-click install into the profile.

Reusable workflows: orchestration scripts saved as files under .dsh/workflows, discovered from the project, and run by name with arguments, so they live in git and can be shared. Four tools: list, show, run, save.

Intervals.icu-only endurance-coaching skill bundle that turns Intervals.icu training data (activities, intervals, streams, training load CTL/ATL/TSB/rampRate, HRV, resting heart rate, sleep in-bed hours, eFTP, power curves) into concrete training decisions: readiness gating, VDOT pace and heart-rate zones, periodized weekly plans written back to the Intervals.icu calendar as events after confirmation, session reviews, race-day pacing and fueling, and weather-adjusted workouts. The bundle registers the endurance-coach skill in the profile, ships deterministic helper scripts, reads an athlete profile created by its init script, and requires the agent to state when data is unavailable instead of inventing it. It needs a separately installed Intervals.icu MCP server for training data and Python 3.10+ for its bundled scripts; with neither it works only from numbers you type in.

Draw.io in chat: describe it, get an interactive diagram in the transcript; 12 Chinese scenario templates; keep editing or undo the same .drawio file; right-sidebar preview. No account, no MCP server, no editor.

Bridges the official COROS MCP server into DeepSeek Harness with a self-contained local stdio-to-HTTP bridge and on-device OAuth login, exposing 34 COROS tools on the current endpoint for sleep, HRV, resting heart rate, recovery, training load, activities and FIT files, plus workout and training-plan CRUD and scheduling.

Decides which MCP server tools each workspace can see, keyed by the session working directory; workspaces without a policy see none.

Wires the local cc-computer-use MCP server into DSH with one process per root session, giving each session its own Xephyr sandbox screen and private AT-SPI bus (requires the cc-computer-use MCP server).

MCP and skill manager as a full sidebar page: add, edit, enable, disable and delete MCP servers, expand a server to its tools and open one for its parameter list, translate skill and tool summaries with a model you pick, and count tool and skill invocations. Also registers an mcp_service tool so the agent can install MCP servers in conversation after an approval prompt.

Connects DeepSeek Harness to the hosted Renor memory MCP server over Streamable HTTP, using a Renor API key from RENOR_API_KEY, so the model can call Renor's recall and save tools. It is a configuration bundle with no code of its own.

Brings the third-party deepseekbase.com index into DSH: eight tools and five DSHIO standard skills (from the @liangjiehappy/dsh-dshio dependency). Search 8,400 plugins with real sandbox install verdicts and risk levels (including red-line warnings where credential access meets network), submit a plugin for a four-stage sandbox test from the conversation, read the verdict, check the remaining free quota, and run a clause-by-clause DSHIO/T 10 conformance check. The skill text is fetched from dshio.org at startup and replaces the bundled copy, so updates need no reinstall. Network: HTTPS to deepseekbase.com and dshio.org, plus an npm registry lookup before a test is submitted; no file reads or writes, no credential access, no shell.

勿忘我：给任何 LLM 用的封存式记忆大脑——艾宾浩斯遗忘、联想召回、可塑性性格、自动总结，附 3D 莫比乌斯记忆星环与脑图。DeepSeek Harness 插件 / MCP / REST