The governed AI coding platform

    Give engineers autonomous coding agents without losing control of models, data, quality, or spend. Start serverless. Deploy in your VPC when ready.

    Get KimchiSee Enterprise Governance
    curl -fsSL https://github.com/getkimchi/kimchi/releases/latest/download/install.sh | bash
    kimchi
    v0.0.26~/project · main
    Try a /ferment workflow

    The agent breaks the task into milestones, self-evaluates its output and delivers with minimal interruptions.

    Skip
    ↑↓ navigate · enter selectkimchi agent
    AkamaiWeWorkRokuCiscoYotpoSwisscom

    Prompt to PR. No babysitting required.

    OSS models

    Built for OSS models.

    • 50+ models. Zero lock-in.
    • Automatically orchestrates models by task type and cost
    • Frontier OSS models via our Minimax partnership
    kimi-k2.7reasoning262k context$4.50/1M output tokens
    minimax-m3execution1M context$1.20/1M output tokens
    nemotron-3-super-fp4execution1M context$0.75/1M output tokens
    Avg. price of proprietary models$12.10/1M output tokens

    /ferment workflow

    Hand off the task. Come back to a finished PR.

    A workflow that allows you to plan upfront, approve once and forget the terminal. The agent builds through each milestone autonomously, and a PR appears when it's done. Not another question.

    /ferment implement a "to do" app
    ✓ plan approved
    ✓ Phase 1: Scaffold project structure0.8s
    ✓ Phase 2: Build task model + local storage1.1s
    • Phase 3: Create UI componentsrunning…
    • Phase 4: Add due dates · tests · open PR

    Code session

    Stay in the loop. Have more control. Make it quick.

    The standard coding agent conversation powered by our open source model orchestration. Interact more closely with the agent and stir it exactly in the direction you want.

    Rename userId to accountId · Auto permissions
    Found 6 more references in other files.
    Want me to update those too?
    ❯ Yes
    No

    One command setup

    Migrate from Claude Code. Keep everything.

    On first run, kimchi setup detects existing Claude Code and OpenCode installations — MCP servers, skills, configuration — and offers to migrate in one prompt.

    ┌ Claude Code configuration found
    │ MCP servers: filesystem, github, ripgrep
    │ Claude Code skills: 4 in ~/.claude/skills
    │ OpenCode skills: 2 in ~/.config/opencode/skills
    ◇ Migrate MCP servers to Kimchi?
    ● Migrate now
    ○ Skip this time

    Evaluator

    Every task is graded before it ships.

    After each milestone, a high-reasoning model scores the output against the original intent, catching regressions, half-done work, and silent failures before they compound. Use Opus, Kimi, or whatever model you trust most for the job.

    evaluator · claude-opus-4-7
    spec match9.2 / 10
    regressions0 found
    edge cases2 flagged
    ✓ milestone passed · proceeding to 4
    evaluation log
    ✓ M1: scaffold — 9.6/10
    ✓ M2: data layer — 9.1/10
    ▲ M3: UI — 2 edge cases
    ○ M4: tests + PR — pending

    VS Code extension

    Kimchi right inside your editor.

    The extension brings the full harness into your IDE with agent sessions, model orchestration, and governance controls without leaving your editor.

    kimchi-panel.tsx
    > Refactor auth middleware
    ✓ Planning phase complete
    ▲ Executing: 3 of 7 steps
    ○ Next: integration tests
    Install from Marketplace
    Kimchi Studio task list

    Kimchi Studio. The harness in a visual workspace.

    All the power of the CLI harness inside a visual workspace. Chat with an agent to set up tasks, then watch them run in parallel. Code mode ships with terminal UI and chat UI; design mode is on the way.

    • Visual task setup. Chat with an agent to define the work
    • Watch tasks run in parallel from one view
    • Code mode: terminal UI + chat UI
    • Desktop app for macOS, Windows, and Linux
    • MCP servers & Slack connector work in Studio
    Explore Studio

    Hand off to a remote workspace. Close the laptop.

    Run /teleport to hand your in-progress session to a persistent cloud sandbox, provisioned from a template image with the tools your task needs. Close the lid and the session keeps running. Reconnect via /remote-sessions or VS Code.

    • Persistent sandbox keeps running when you disconnect
    • Resume from the CLI, the Kimchi console, or VS Code
    • Browse workspaces and sessions with /remote-sessions
    • Raw SSH shell anytime with /terminal
    Explore Remote Sessions
    Kimchi remote session terminal
    Budget controls dashboard

    Spend limits for total ease of mind.

    See exactly who, what, and where your AI spend is going. Per-developer, per-team, per-model, per-tag - in real time. Hard budget caps at every level: user, team, API key.

    • Per API key set limits for CI/CD pipelines, bots, or contractors
    • Per user every developer has their own budget with real-time visibility
    • Per team finance gets a clean number; engineering keeps shipping
    "Kimchi is fundamentally changing how our engineering team thinks about AI-assisted development. Auto router keeps the quality of a single frontier model while cutting token costs dramatically."
    Dekel ShavitDekel ShavitSr. Director of Engineering

    Build with freedom. Start in 60 seconds.

    Install the CLI, run /ferment, and hand off to the agent. No lock-in, no waiting list, no credit card.

    Get KimchiStar on GitHub