Kilo Code
Open source AI coding agent for VS Code, JetBrains, or the command line.
- Stars
- 27.8k
- Forks
- 3.3k
- Licence
- MIT
- Language
- TypeScript
- Last update
- 4 days ago
- Type
- AI model / agent
What it is
Kilo Code is an AI coding agent that works inside VS Code, JetBrains IDEs, or a terminal CLI. You describe a change in plain language and the agent writes or edits code across multiple files, can run commands, and can switch between many AI models. According to the README, you pay the model provider's rate with no markup and can start without an API key.
- Generate or edit code across several files from a natural-language request
- Ask questions about a codebase in Ask mode without changing files
- Use Plan mode to outline an implementation before code is written
- Run `kilo run --auto` in a CI/CD pipeline to run tests and fix failures
- Set up automated AI code reviews on pull requests via the Kilo web app
Great for
- Developers who want an AI coding agent inside VS Code or JetBrains
- People who want to choose among many AI models for different tasks
- Teams wanting automated AI checks in CI/CD pipelines using the CLI
- Developers who prefer open source tools they can inspect and modify
Not ideal for
- People who want a fully offline tool, since the README describes use of model providers and a cloud service
- Non-programmers, since it is built for editing and running code
- Anyone running untrusted automation, because --auto approves prompts unless a rule denies them
How to get it
- 1
Install with a package manager
You'll need: Node.js or a package manager such as npm, pnpm, or bun (for the npm-based CLI install) · VS Code or a JetBrains IDE (for editor use)
Install the CLI with npm
npm install -g @kilocode/cliInstall the CLI with curl
curl -fsSL https://kilo.ai/cli/install | bash - 2
Run it
- Install the extension in VS Code or the plugin in JetBrains, or install the CLI
- Run `kilo` in any project directory to start the CLI
- Pick an agent such as Code, Plan, Ask, or Debug depending on the task
- Describe the change you want in plain language and review the result
- 3
Get the most out of it
- Use Plan mode before large changes to review the approach first
- Use Ask mode to explore an unfamiliar codebase without edits
- Switch models mid-task to balance cost, speed, and reasoning
- Add MCP servers or agents from the Kilo Marketplace to extend capabilities
- Configure permission rules before using `--auto` in CI/CD
Downloads
Strengths
- Open source under the MIT license, so the code can be read and reused
- Works across VS Code, JetBrains, and the command line
- Lets you switch between many models mid-task
Trade-offs
- Model access and pricing depend on the provider and Kilo's hosted service, which the README describes only briefly
- Autonomous mode skips permission prompts unless rules block actions, so it needs careful setup
- The README is truncated, so some details like the full feature set and limits are not confirmed here
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| kilocode (this) | 27.8k | AI model / agent | beginner | 4 days ago | Developers who want an AI coding agent inside VS Code or JetBrains |
| opencode | 224k | AI model / agent | intermediate | 4 days ago | Developers who want an AI agent that works inside their terminal and existing projects |
| GPT_API_free | 45.3k | Other | beginner | 7 days ago | Students and researchers doing non-commercial experiments with LLM APIs |
| watermarks-remover | 25.9k | AI model / agent | intermediate | 5 days ago | Developers who use AI coding agents and want written files scrubbed of provenance metadata |
| LibreChat | 46.6k | App | intermediate | 4 days ago | Teams that want one shared chat interface across several AI providers |
| one-api | 37.3k | App | intermediate | 10 months ago | Developers who want one OpenAI-compatible endpoint for several model providers |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.