AI Agent Hub
Back to plugins
🖥️

dsh-mcp-ui

Client Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install quan-v/dsh-mcp-ui

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install quan-v/dsh-mcp-ui in your terminal to install the plugin; the source repository is at https://github.com/quan-v/dsh-mcp-ui .

About this plugin

Connecting MCP servers to DeepSeek Harness used to mean opening settings.yaml, hand-typing stdio commands, URLs, and headers, then restarting the process. dsh-mcp-ui turns that into a two-step ritual: open the settings card, fill in a few fields, hit save. The server is live and the model can call its tools immediately—no restart required.

It covers both stdio (local process) and streamable-http (remote URL) transports with full field support: serverName, command, args, env, cwd, url, headers, tool-call timeout, startup-failure policy, and auto-reconnect parameters. The moment you save, the plugin mounts or unmounts the corresponding dsh-mcp-client instances on the host side for instant hot sync. Sensitive values can be referenced as environment variables so they never appear in plaintext config.

Ideal for individual developers or small teams already working in the DSH Web console who want a zero-terminal workflow for managing MCP server connections—especially when juggling local and remote endpoints or consolidating tool registration into a single settings panel.

Use Cases

  • Add, edit, or remove local and remote MCP server connections from the DSH Web settings panel
  • Swap between stdio and streamable-http endpoints without restarting the process
  • Consolidate MCP tool registration into a single plugin settings card

Best For

  • Developers working in the DSH Web console who prefer a zero-terminal MCP workflow
  • Small teams that frequently debug local stdio servers or remote HTTP endpoints
  • DSH users who want save-and-go configuration without hand-editing settings.yaml