LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
PluginRisingofficial publisher

Kilo Code

Plugin : A set of capabilities installed into an AI application or a host · requires vscode

Open source coding agent, successor to Roo Code, with modes, MCP, and model routing.

The continuation of Roo Code, actively maintained and available across multiple surfaces.

Overall score
8.3 /10
Security: one caveat, see the card
Utility
8.0
Reliability
10.0
Security
One caveat
Ease
1 click · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Kilo Code is aimed at developers who want an AI coding agent directly in their working environment, without being locked into a single model provider or paying a markup on API calls. It addresses the problem of opaque costs and lack of transparency found in proprietary code assistants.

In practice, Kilo Code installs as an extension in VS Code or JetBrains, or runs from the command line. It reads your codebase, generates code across multiple files from natural language instructions, executes terminal commands, and can control a browser. It offers 500+ models at direct provider pricing with no markup, and allows switching models mid-task.

The tool suits solo developers or teams who want to keep control over their AI spending, freely choose their models, and audit agent behavior through MIT-licensed open source code. It also targets teams looking to integrate an agent into CI/CD pipelines via autonomous mode.

That said, Kilo Code requires some familiarity with development environments. Users looking for a fully turnkey tool with no configuration or API key management will find the onboarding more demanding, even though a Kilo account allows getting started without a personal API key.

What it offers

  • ·Five built-in specialized agents: Code, Plan, Ask, Debug, and Review, each suited to a specific type of task.
  • ·Access to 500+ AI models at direct provider pricing with no markup, with the option to bring your own API key.
  • ·Automatic model routing via Auto Model mode, which selects the model based on budget and task type.
  • ·Inline autocomplete with ghost-text suggestions and tab-to-accept.
  • ·Terminal and browser control to run commands and automate web actions.
  • ·MCP protocol support with a marketplace to connect MCP servers and extend agent capabilities.
  • ·Autonomous mode for CI/CD pipelines via the kilo run --auto command, with no confirmation prompts.
  • ·Automated code review on pull requests and cloud agents accessible from the web, without a local machine.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
vscode, jetbrains, cli
Host
vscode
Licence
Apache-2.0
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Your data
Local or cloud depending on use
Maintenance
Active
Last release
23 October 2026

Links and repository checked on 25 October 2026 · added on 1 April 2025

How to install Kilo Code

1. Open VS Code and go to the Extensions tab (Ctrl+Shift+X). 2. Search for "Kilo Code" and click Install, or use the direct link from the VS Code Marketplace. 3. Create an account at kilo.ai to access models without a personal API key, or enter your own API key in the extension settings. 4. Select the agent suited to your task (Code, Plan, Ask, Debug, or Review) in the Kilo Code panel. 5. Always review the changes proposed by the agent before accepting them, especially terminal commands, as the tool has read and write access to your files.
npm install -g @kilocode/cli

In the same category

Whole category Code →