DSH Plugins · Community project

dsh-TUI

Add a full-screen terminal interface to DeepSeek Harness with streaming thoughts, tool cards, context progress, TPS, session actions, and rollback.

2.2k StarsMITTypeScriptUpdated today
01

Project overview

Add a full-screen terminal interface to DeepSeek Harness with streaming thoughts, tool cards, context progress, TPS, session actions, and rollback.

dsh-TUI showing work state, context progress, and TPS in the terminal
The dsh-TUI working status line shown in the repository README.View repository image
02

Core capabilities

01

Stream model thoughts and replies in the terminal

Stream model thoughts and replies in the terminal. This capability is documented in the repository README and still requires verification against the target DSH version and profile.

02

Render execution state as tool cards

Render execution state as tool cards. This capability is documented in the repository README and still requires verification against the target DSH version and profile.

03

Show context progress and TPS

Show context progress and TPS. This capability is documented in the repository README and still requires verification against the target DSH version and profile.

04

Resume sessions and roll back with double Escape

Resume sessions and roll back with double Escape. This capability is documented in the repository README and still requires verification against the target DSH version and profile.

03

Installation and usage

Install in DeepSeek Harness, reload the target environment as documented, and verify it with a low-risk scenario.

AI AGENT INSTALL

Let an AI Agent install it

Send this prompt to Codex, Claude Code, or another AI agent that can work with your local environment.

Help me install dsh-TUI from https://github.com/ccch1mneyyy/dsh-TUI. Read the README, license, and installation files first. Confirm the current DeepSeek Harness version, target profile, and required dependencies. Use the repository's current command `npm install -g @deepseek-ai/dsh @deepseek-harness-tui/dsh-tui`. Explain the profile, paths, and permissions that will change before running it. Follow the repository verification steps and report commands, changed locations, and visible results. Ask before requesting credentials, enabling extra network access or build scripts, overwriting files, or expanding permissions.
01Before you start
  • An interactive terminal TTY
  • The official dsh CLI
  • pnpm 10 or later
  • DEEPSEEK_API_KEY for model execution
02Copy the install command or configuration
npm install -g @deepseek-ai/dsh @deepseek-harness-tui/dsh-tui
03Complete the setup steps
  1. 1
    Check the environment and target profile

    An interactive terminal TTY; The official dsh CLI; pnpm 10 or later; DEEPSEEK_API_KEY for model execution. Record the current configuration and installed plugins before changing anything.

  2. 2
    Run the current repository command

    Run the following command. `npm install -g @deepseek-ai/dsh @deepseek-harness-tui/dsh-tui` Stop and ask before enabling build scripts, providing credentials, or overwriting files.

  3. 3
    Reload and verify

    Run `dsh-tui`, confirm that the dsh-tui profile initializes, and check for the whale header, status line, and an interactive session prompt.

How to verify the setup

Run `dsh-tui`, confirm that the dsh-tui profile initializes, and check for the whale header, status line, and an interactive session prompt.

Before using it
  • Requires a real TTY rather than a plain log pipe
  • Initial profile setup depends on pnpm
  • Terminal shortcuts can be intercepted by the operating system or terminal app
04

Use cases

SCENARIO 01

Use DSH from a server or remote terminal

Start with a minimal dsh-TUI scope in this scenario, then expand only after checking output, permissions, and compatibility.

SCENARIO 02

Run long keyboard-driven coding sessions

Start with a minimal dsh-TUI scope in this scenario, then expand only after checking output, permissions, and compatibility.

SCENARIO 03

Monitor context use and tool execution state

Start with a minimal dsh-TUI scope in this scenario, then expand only after checking output, permissions, and compatibility.

05

Assessment

This assessment is based on the repository README, installation instructions, license, and maintenance metadata. Adds a terminal-first DSH workflow; Combines context, throughput, and tool state in one view; Supports session resume and rollback. Requires a real TTY rather than a plain log pipe; Initial profile setup depends on pnpm; Terminal shortcuts can be intercepted by the operating system or terminal app. No local installation or long-term use is claimed.

Why it may be useful

  • Adds a terminal-first DSH workflow
  • Combines context, throughput, and tool state in one view
  • Supports session resume and rollback

What to know first

  • Requires a real TTY rather than a plain log pipe
  • Initial profile setup depends on pnpm
  • Terminal shortcuts can be intercepted by the operating system or terminal app
06

README

dsh-TUI


Overview

Add a full-screen terminal interface to DeepSeek Harness with streaming thoughts, tool cards, context progress, TPS, session actions, and rollback. Add a full-screen terminal interface to DeepSeek Harness with streaming thoughts, tool cards, context progress, TPS, session actions, and rollback.

Getting started

  • Install in DeepSeek Harness, reload the target environment as documented, and verify it with a low-risk scenario.
  • Add a full-screen terminal interface to DeepSeek Harness with streaming thoughts, tool cards, context progress, TPS, session actions, and rollback.
  • The repository documents these main capabilities: Stream model thoughts and replies in the terminal, Render execution state as tool cards, Show context progress and TPS, Resume sessions and roll back with double Escape.
  • Run `dsh-tui`, confirm that the dsh-tui profile initializes, and check for the whale header, status line, and an interactive session prompt.

Configuration

npm install -g @deepseek-ai/dsh @deepseek-harness-tui/dsh-tui
Read the complete README on GitHub