Not a vibe coder.
It's an
engineering tool.

Built for engineers who ship production code — not prompt-and-pray demos. Plan, write, review, and test with full control over every change, using the models you already trust.

Works with the models you already pay for

Anthropic · OpenAI · Google · GitHub Copilot · DeepSeek · Groq · Ollama · xAI · OpenRouter · Mistral

Multi-Agent

Parallel agents. No conflicts.

Plan, Code, Pilot, Review, Advisor — specialized agents running side by side so big tasks finish faster.

Five Kodein sub-agents — Plan, Code, Pilot, Review, Advisor — each running a different model side by side

Diagrams

Diagrams, rendered in chat.

Ask for a flowchart, sequence, Gantt, or timeline and Kodein draws it inline — 30 Mermaid diagram types rendered natively in pure Dart. No browser, no network, no round-trip.

A Mermaid flowchart of Kodein's agent loop — user prompt to agent loop to a model-decides diamond that either runs a tool and loops back or emits the final answer — rendered natively in chat

Code Search

A local semantic index.

Kodein indexes your whole codebase on-device — files, symbols, and vectors — so it finds code by meaning, not just keywords. Embeddings run on your own GPU.

Kodein code index card showing files, symbols, vectors, and on-GPU embeddings
Kodein quota cards showing live usage for a Claude subscription and a GitHub Copilot plan

Sign In

Bring your own plan.

Sign in with your existing Claude, ChatGPT, Gemini, or Copilot subscription — or drop in an API key. Kodein tracks your live usage and limits so you always know where you stand.

Usage

Every token, every dollar.

Live context usage, cache savings, and exact cost stay in view — never a surprise bill.

Kodein token usage panel showing input, output, cache, and cost per turn
Kodein MCP Servers panel listing connected servers — filesystem, github, postgres, playwright, slack — with live status and tool counts

Extend

Connect anything.

Plug in MCP servers, language servers, and a built-in browser. Kodein speaks the protocols your stack already uses.

Memory & Skills

It learns your project.

Kodein remembers decisions across sessions and lets you save reusable skills it can run on demand.

Kodein Memory and Skills sidebars listing remembered project facts and reusable skills
Kodein permission request card asking to allow a command, with allow-scope dropdown and Allow, Deny, and Interrupt actions
Running in a sandbox — file writes and commands are scoped to your project.

Permissions & Sandbox

Nothing runs without your say-so.

Every command and file write asks first — allow once, always allow, or deny. Work runs sandboxed and scoped to your project, so the agent can move fast without going off the rails.

Multi-Review

Review with a panel of models.

Choose how many reviewers and which model each one runs. They review your diff independently, then a supervisor merges the verdicts — so no single model's blind spot slips through.

Kodein multi-reviewer panel: pick how many reviewers and which model each one runs
Kodein review panel showing a syntax-highlighted diff with per-hunk Accept and Reject buttons

Review

Every change, every hunk.

Review what the agent changed and accept or reject per hunk or line — even run a panel of models to review for you.

Multi-Advisor

A second opinion — or three.

Before you commit to an approach, fan the question out to multiple advisors on different models and compare their takes side by side — no edits, just perspective.

Kodein multi-advisor panel: fan a question out to multiple advisors on different models
25+ AI providers
8 specialized agents
100% local execution
0 telemetry by default

Specialized Agents · Multi-Model

One task, many minds.

Purpose-built agents with their own tools and prompts — and for the calls that matter, a whole panel of models instead of a single opinion.

Code
Orchestrator
Plan
Analyst
Ask
Q&A
Pilot
Delegator
Explore
Researcher
Worker
Executor
Smart
Reasoner
Review
Verifier

Any Model

Use the plan you already pay for.

Sign in with your subscription — or drop in an API key. Same agent, every provider, no markup.

Anthropic
Claude
SSO · Claude Pro / Max
Sign in with your Claude subscription — no per-token billing.
API key · bring your own
OpenAI
ChatGPT · GPT-5
SSO · ChatGPT Plus / Pro
Sign in with ChatGPT — drives GPT-5 through the Codex backend.
API key · bring your own
Google
Gemini
SSO · Gemini Code Assist
Sign in with Google — use your Gemini Code Assist plan.
API key · bring your own
GitHub Copilot
Multi-model
SSO · Individual / Business / Enterprise
Sign in with GitHub — one Copilot seat, every Copilot model.
API key · bring your own
Also supported
DeepSeek
Groq
Ollama
LM Studio
xAI
OpenRouter
Cohere
Mistral
Together
Fireworks
Vertex AI
Qwen
Nebius

Any OpenAI-compatible endpoint works too — point Kodein at your own inference server.

Platforms

Native everywhere.

No Electron. No webview.

Available now

macOS

Native NSWindow tab groups. Single-instance, GPU-accelerated.

Available now

Linux

Native GTK. Terminal support, filesystem access, Ollama local inference.

Available now

Windows

Win32 with Mica/Acrylic. CUDA + Vulkan for local GPU inference.

Coming soon

VS Code

Full Flutter UI inside VS Code via gRPC-Web. Same agent, same tools.

Download

Get Kodein for your platform.

Always the latest release. Pick your operating system and CPU architecture.

Platform x64 / Intel arm64 / Apple Silicon
macOS
Download Download
Windows
Download
Linux
Download Download

Looking for a specific version or the changelog? Browse all releases.