dsh-superagent

dsh-superagent is a research prototype from the DeepSeek Harness team, published under the deepseek-ai GitHub organization. It is a proactive coding agent for a single GitHub repository, packaged as a bundle for DeepSeek Harness. The README states it is not an official DeepSeek product and should be treated as research code.

What It Is

The agent listens to GitHub and, optionally, Feishu. When an allowed person asks it for work, it opens a work item, works on it in its own agent session and directory, and answers in the original conversation, for example with a pull request. A web panel shows which items need a person.

Working Principle

The project is framed as a way of working with an agent: take as little human attention as possible and give as much value as possible. According to the README, the agent acts only when asked or when a rule the team wrote allows it, and speaks only when someone must act or can use a result. The philosophy document explains this principle and what it asks of the operator.

Setup Path

Setup has many steps, so the README suggests asking a coding agent to read SETUP.md and set up the project interactively. The agent asks one short question at a time, shows the full plan once, then installs and configures everything. Numbered documents cover the domain model, authorization, execution, prompts, the panel, deployment, architecture, and a cookbook for adding a Source.

Current Status

The README says the published code was adapted from the team's internal version and differs from the implementation described in the article Exploring Super-agent Design: Towards a More Autonomous Agent System. Some designs in that article are future directions and are not implemented. It may contain bugs and should be reviewed before use in other environments.