Getting Started

Quick Start

Get up and running in 3 steps. Come back to the rest of the doc when you're ready.

1. Install

curl -fsSL https://github.com/getkimchi/kimchi/releases/latest/download/install.sh | bash

2. Get your API key

Go to https://app.kimchi.dev/settings and generate an API key.

3. Pick your path

"I want to try the Kimchi coding agent"

kimchi
# You'll be prompted for your API key on first run.
# Then just describe a task: "add tests for the auth module"

"I already use Claude Code / Cursor / OpenCode / Cline — just point it at Kimchi"

kimchi setup
# Interactive wizard detects your tools and configures them automatically.

That's it. Read on if you want to understand what's happening under the hood.

What Is a Coding Harness?

A coding harness is an AI agent that can write, read, edit, and reason about your code autonomously. You describe a task in plain English — "add error handling to the API routes", "refactor this module to use async/await", "write tests for the auth service" — and the agent executes it: reading files, writing code, running commands, checking its own output.

This is different from AI code completion (Copilot-style inline suggestions). A harness handles multi-step tasks with minimal hand-holding. You don't autocomplete one line at a time — you hand over a job and the agent does it.

Why Kimchi Instead of Claude Code or Others?

The key differentiator: orchestrated multi-model routing.

Kimchi doesn't just swap one model for another. The harness runs an orchestrator that classifies your task, breaks it into phases, and assigns each subtask to the right model from its roster. A planning step gets a reasoning-capable model; a bulk code-generation step gets a fast, high-throughput executor. You don't pick the model — the harness does.

Two Ways to Use Kimchi

Mode 1: The Kimchi Harness (kimchi)

Run the full Kimchi coding harness:

kimchi

This launches kimchi — a terminal-based coding agent with:

• Multi-model orchestration (on by default): the agent classifies your task, assigns phases, delegates subtasks to specialised subagents
• Phase tracking: work is tagged as explore → plan → build → review → research for analytics and cost attribution
• Cost tagging: every request is tagged by model and phase; add custom tags for project/team attribution
• Session persistence: sessions (including subagent runs) are saved to disk and fully recoverable

When to use: New tasks, complex tasks, anything where you want the full autonomous experience.

Mode 2: Configure Your Existing Tool (kimchi setup)

Already using Claude Code, Cursor, OpenCode, Cline, or another tool? Keep your workflow and just swap the backend:

kimchi setup

The interactive wizard will:

• Ask for your API key and validate it
• Detect installed AI tools on your machine
• Let you pick which tools to configure
• Ask if the config should apply globally or per-project
• Write the config files

When to use: You're happy with your current tool's UX and just want it to run on Kimchi's models instead of Anthropic/OpenAI. You're mainly using AI for code autocompletion and need the UI.

Supported tools:

Using the Harness: Key Features

Multi-Model Orchestration

By default, kimchi runs in multi-model mode: a main orchestrator agent classifies each task and delegates subtasks to the best-fit model from the roster. This happens automatically — you just describe the task.

To turn orchestration off (single model, no delegation):

kimchi --multi-model=false

Or toggle it during a session with alt+tab (shown in footer: multi-model: on / off).

When to turn it off: Simple one-shot tasks where you want direct, fast output without orchestration overhead.

Phase Tracking

The harness tracks which phase of work it's in and tags every LLM request accordingly. This is automatic — the orchestrator sets the phase as it works.

Phases appear in the footer (e.g. ↳ build) and are included in every request tag for cost reporting.

Model Switching

To manually switch models mid-session:

/model

This is an escape hatch — in normal orchestration mode, the harness routes automatically. You can also toggle it during a session with ctrl+p.

When working in multi-model mode the selected model is responsible for the orchestration of the prompt. For best experience we recommend using kimi-* family models.

Quick Reference

# Install
curl -fsSL https://github.com/getkimchi/kimchi/releases/latest/download/install.sh | bash
# Launch the harness
kimchi
# Configure an existing tool
kimchi setup
# Update to latest
kimchi update
# Show version
kimchi version
# Debug mode
kimchi --debug

When to Use Which Mode

Feedback & Issues

File issues at github.com/getkimchi/kimchi or ping the team on Discord here.

Don't forget that Kimchi Code is an open source project so feel free to help us extend it if you have any ideas.