Kilo Code

Overview

Kilo Code is an open-source AI coding agent that operates natively across VS Code, all JetBrains IDEs, CLI environments, and cloud agent workflows. It is engineered to help developers write, debug, refactor, and review code while automating multi-step tasks across the entire repository. As a fully MIT-licensed project, Kilo Code offers complete source code transparency and auditability, ensuring that every prompt and interaction remains visible and under the user's control. With support for over 500 hosted AI models and local model integration, Kilo Code provides a flexible and model-agnostic foundation for AI-assisted software development.

The platform gives developers direct access to a broad spectrum of AI models, including Claude, GPT, Gemini, DeepSeek, and local models served through Ollama or LM Studio. Users can bring their own API keys and pay at exact provider rates with zero markup, or use pay-as-you-go credits that work across every surface — IDE extensions, CLI, Cloud Agents, and App Builder. Kilo Code supports VS Code and all VS Code-based editors such as Cursor forks and Windsurf, as well as every major JetBrains IDE, including IntelliJ IDEA, PyCharm, WebStorm, GoLand, Rider, RubyMine, and CLion. The CLI can be installed with a single command, and a single sign-in synchronizes settings and credits across all environments.

In addition to its core coding-agent capabilities, Kilo Code extends to team workflows and code review automation. Its architecture is built around the principle of model freedom; developers are never locked into a proprietary model or vendor. The open-source codebase allows enterprises to inspect and audit the tool's behavior, and the transparent pricing model ensures that costs align directly with model usage. Cloud Agents enable autonomous execution of coding tasks in remote or server-side environments, while the App Builder facilitates rapid prototyping and application creation directly from natural-language instructions.

Adopted by more than 3 million developers and backed by a 26.9k-star GitHub community, Kilo Code represents a growing ecosystem of developer-first AI tooling. The product is free to install and open source under the MIT license, with no mandatory subscription and no fixed seat price for individual developers. This makes it an ideal choice for independent developers, open-source contributors, development teams, and enterprises seeking a transparent, high-performance, and cost-effective AI coding agent that fits seamlessly into existing workflows.

Core capabilities

500+ Hosted AI Models

Direct access to a vast model library including Claude, GPT, Gemini, DeepSeek, and more from your IDE, CLI, or cloud agent.

Bring Your Own API Keys

Use your own API keys at provider rates with zero markup, ensuring full pricing transparency and no vendor lock-in.

Local Model Support

Connect to local models via Ollama or LM Studio for offline, private, or fully controlled development environments.

Multi-Environment Coverage

Works across VS Code, VS Code-based editors, all JetBrains IDEs, CLI, and Cloud Agents for a unified workflow.

Open-Source and MIT-Licensed

Fully auditable source code with visible prompts and no closed black-box behavior, giving teams complete trust.

Multi-Step Task Automation

Automates complex workflows across a repository, including writing, debugging, refactoring, and code review.

Unified Credit System

Pay-as-you-go credits work across every surface — IDE, CLI, cloud, and app builder — with no subscription required.

Team Workflows and Code Reviews

Supports collaborative review processes and team-level automation to streamline development operations.

App Builder

Rapidly build applications from natural-language prompts, either locally or in the cloud.

Seamless Sync

Sign in once to synchronize settings, credits, and project state across all supported environments.

Use cases

Automating multi-step refactoring tasks across a large repository

Conducting AI-assisted code reviews within team workflows

Debugging and fixing errors in complex codebases

Building complete applications from natural-language prompts via App Builder

Running autonomous coding agents in cloud or CI environments

Integrating local LLMs for privacy-sensitive or offline development

Managing collaborative development processes and team-level automation