Our take
- Activity
- 90
- Community
- 93
- Issues
- 76
- Pull requests
- 75
+348 stars in the last 29 days
- Releases more often than 93% of the projects we track
- More contributors than 78% of the projects we track
- Merges more pull requests than 78% of the projects we track
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
27,862Forks
3,336Change since Nov 4, 2026: +9 stars·+1 forks
Overview
Kilo Code brings an AI coding agent to VS Code, JetBrains IDEs, and the CLI. It is intended for developers who want AI assistance within their existing coding workflow, from asking questions about a codebase to making changes and running commands.
The project combines coding and planning agents with access to multiple models and tools such as terminal and browser control. It may suit teams or individual developers who want to switch models or work across different development environments; it is less suited to users seeking a standalone model to run locally.
Key Features
- Works through VS Code and JetBrains integrations as well as a CLI.
- Provides Code, Plan, Ask, and Debug agents for different tasks.
- Generates and edits code across multiple files from natural-language instructions.
- Offers inline autocomplete and self-checking of agent work.
- Can use terminal and browser tools to run commands and automate web tasks.
- Supports switching among 500+ models during a task, according to the README.
- Includes an autonomous CLI mode for CI/CD, which can approve permission prompts unless a rule denies the action.
Use Cases
- Developers can ask the Ask agent to explain a codebase without modifying files.
- Engineers can use Plan to outline an implementation before editing begins.
- Developers can delegate a feature or bug fix to the Code or Debug agent while working in an IDE or terminal.
- Teams can use the CLI's autonomous mode to run a task in a trusted CI/CD environment, with care around permissions.
Getting Started
Install the CLI with npm and run it in a project directory:
npm install -g @kilocode/cli
kilo
For IDE installation, other installation options, configuration, and usage details, see the README and documentation.
Alternatives
- opencode: OpenCode focuses on terminal and desktop coding workflows, while Kilo Code also integrates with VS Code and JetBrains IDEs.
- pi: Pi is an extensible TypeScript agent toolkit and CLI, while Kilo Code is a coding agent offered across IDEs and the command line.
- Codebuff: Codebuff edits codebases through a multi-agent terminal workflow, while Kilo Code also supports IDE integrations and planning, debugging, and codebase-question modes.
- Qwen Code: Qwen Code is a command-line agent adapted for Qwen models, while Kilo Code supports multiple models and IDE integrations.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| kilocode | TypeScript | MIT | 27.5k | Not checked yet |
| opencode | TypeScript | MIT | 212k | Active |
| Pi | TypeScript | MIT | 117k | Active |
| freebuff | TypeScript | Apache-2.0 | 13.1k | Active |
| qwen-code | TypeScript | Apache-2.0 | 28.6k | Active |
Considerations
- The CLI and IDE integrations provide an agent interface, not a local model; model access depends on the available providers and Kilo's service.
- Autonomous mode can approve permission prompts, so the README recommends using it only in trusted environments.
- The repository identifies the CLI as a fork of OpenCode, enhanced for the Kilo platform.
- The repository is MIT-licensed, but using AI models may involve provider pricing; the README describes pricing as the model provider's rate with no markup.
Found this useful?
Share it with someone who would like kilocode.