Quick Answer
DeepSeek Harness (dsh) is DeepSeek AI's open-source agent harness — an agent framework built on an "everything is a plugin" architecture, powered by Cordis and released under the MIT license. It is currently in developer preview and iterating fast, so expect compatibility-breaking changes.
Installation needs only Node.js. The fastest way to run it:
npx @deepseek-ai/dsh web
This pulls the package and starts the Web UI at http://127.0.0.1:3080 by default. To run it on DeepSeek V4 Pro / Flash, point the API key and base URL at Moyi API. Details below.
October 31, 2026: DeepSeek open-sourced Super-agent (dsh-superagent), a research-prototype coding agent that runs as a DeepSeek Harness plugin and works on one GitHub repository.
What DeepSeek Harness Is
First, what it is and isn't. The industry has converged on a clean formula:
Agent = Model + Harness
The model is the brain — the weights that predict tokens. The harness is everything else: the tools the agent can call, the filesystem and shell it can touch, how sub-agents pass information to each other, how context is stored and retrieved, and — critically — when execution stops. A bare model is bad at memory, code execution, and tool calling on its own; the harness turns a chat model into a worker that can actually edit code and run commands.
DeepSeek Harness's defining trait is its architecture: everything is a plugin. The filesystem, the shell, the model adapter, the Web UI, and sub-agents are all swappable Cordis plugins. That gives you two things: you can trim an agent's capability surface (give it only terminal + file ops, no network), and you can swap one provider (e.g. move the model from the official API to Moyi API) without touching anything else.
Prerequisites
- Node.js — any recent version (for
npx). Building from source additionally needspnpmandgit. - A working model backend: a DeepSeek API key, or a Moyi API key (
sk-...).
Method 1: One command via npm (recommended)
npx @deepseek-ai/dsh web
The first run downloads the package, then the Web UI starts. Open http://127.0.0.1:3080 for the console. This is the lowest-friction, everyday path — no clone, no build.
To make the agent actually do work, configure the model backend. The quickest way is two environment variables:
export DEEPSEEK_API_KEY="sk-your-key"
export DEEPSEEK_BASE_URL="https://api.moyi-api.com/v1"
Put these in your shell profile (~/.zshrc or ~/.bashrc), or export them before launch, and dsh sends model requests to Moyi API instead of DeepSeek's official API. We tested this with dsh 0.2.1-rc.3 on November 2, 2026: dsh keeps its built-in DeepSeek provider and sends Anthropic-format requests to https://api.moyi-api.com/v1/messages. The default model is deepseek-flash (DeepSeek V4.1 Flash), and you can switch to deepseek-v4-pro in the Web UI's model picker. To list other Moyi API models or change the default model, add Moyi API as a provider in ~/.dsh/settings.yaml — see Connect DeepSeek Harness to Moyi API.
New accounts start with a $0 balance — top up before your first request. Top up from $5. If a request returns 402 (400 on the OpenAI format) with "wallet balance is insufficient", your balance is empty — your base URL and key are fine. Top up and retry.
Method 2: Build from source
Use this when you want to modify, track the latest code, or contribute to the plugin ecosystem:
git clone https://github.com/deepseek-ai/deepseek-harness.git
cd deepseek-harness
pnpm install
pnpm run build
pnpm dsh web
pnpm dsh web and npx @deepseek-ai/dsh web are the same command — the former uses local source, the latter the published npm build. Source builds make sense during developer preview, since main often runs a few minor versions ahead of the release.
After Launch: What's in the Web UI
http://127.0.0.1:3080 is a local web console with these capabilities:
- Session management — create / switch / rename sessions, run several in parallel.
- Model selection — switch between configured models (e.g.
deepseek-v4-provsdeepseek-v4-flash). - Tool panel — the tools the agent can actually call: bash, file read/write, sub-agents, web search, and more.
- Goals — break long tasks into stateful goals, pause / resume / mark complete.
A common misconception is treating dsh as "yet another chat window." It is really the front-end of an agent runtime: you give a goal, the agent loops over tools until done, and the UI is your window for observation and intervention.
A Minimal Working Example
After setting DEEPSEEK_API_KEY / DEEPSEEK_BASE_URL, give the agent this task in the Web UI:
List the current directory, find the README, summarize its first paragraph in one sentence, and write it to /tmp/summary.txt
dsh calls bash (list files, read the file) and the file tools in sequence, then reports back. That is the most direct demonstration of "the harness turns a model into a worker."
Using It with DS Bot, a Community Personal-Assistant Plugin
DS Bot (the repository was first named DeepSeek-Bot) is a DeepSeek Harness plugin built by a community developer and released around October 11, 2026. It turns dsh into a team of bots with long-term memory that can message each other, hold group chats and split up work. It is not an official DeepSeek product.
The plugin uses the models you set up in dsh. Its README says to install dsh and set up at least one model first, because DS Bot uses the models from dsh. So once Moyi API is configured in dsh as above, add the plugin:
npx @deepseek-ai/dsh@0.2.1-rc.4 plugin --profile web add ds-bot
npx @deepseek-ai/dsh@0.2.1-rc.4 web
This needs Node.js 22.19 or later and pnpm. DS Bot also has a desktop app with its own copy of dsh; there you set up the model in DS Bot's own settings. These steps follow DS Bot's README as of November 2, 2026; we have not installed the plugin ourselves.
FAQ
Is dsh tied to DeepSeek's official API?
No. dsh is model-agnostic: DEEPSEEK_BASE_URL points its built-in DeepSeek provider at another address, and ~/.dsh/settings.yaml can add any OpenAI-compatible provider. With Moyi API it runs the DeepSeek V4 family, and can route Claude, GPT, and Gemini from a single key.
What does "developer preview" mean?
Interfaces and config will still change, so don't depend on a specific API detail in production. For solo developers, learning agent architecture, and cheap agent experiments, it is already fully usable.
I installed it but the agent won't run / I get auth errors?
Nine times out of ten the env vars aren't applied. Run echo $DEEPSEEK_API_KEY to confirm the key is set; make sure DEEPSEEK_BASE_URL includes /v1 (dsh appends /messages to it); and confirm the key starts with sk- with no stray whitespace.
To run DeepSeek V4 Pro in dsh, register Moyi API, grab a key, set DEEPSEEK_BASE_URL as above and switch to deepseek-v4-pro in the model picker.