Wedoany.com Report on Feb 6th, Kilo, an AI coding startup focused on remote collaboration, recently launched Kilo CLI 1.0. This command-line tool has been completely refactored and can connect to over 500 different underlying AI models, including open-source solutions from companies like Alibaba.

In an interview, Kilo CEO Scott Breitenother stated, "The current developer workflow experience is fragmented. Engineers sometimes use the command line, sometimes operate in VS Code, and sometimes launch agents from Slack. People shouldn't have to constantly switch between different tools." He believes Kilo CLI 1.0 is designed precisely for these diverse work scenarios.

Built under the MIT open-source license, the tool offers three main operating modes: Code Mode for quick generation and multi-file refactoring, Architect Mode for high-level planning, and Debug Mode focused on system issue diagnosis. To address the problem of AI agents losing context between sessions, Kilo introduced a "Memory Bank" feature that saves work states through structured files.

In terms of extensibility, Kilo CLI 1.0 supports the Model Context Protocol standard and can integrate with external tools. Pricing adopts a transparent "Kilo Pass" subscription system, with precise billing based on vendor API rates, offering three tiers: Starter, Pro, and Expert. A limited-time promotion is currently available, providing new users with extra credits for their first two months.

The release of Kilo CLI 1.0 puts it in competition with terminal tools like Claude Code and Goose. Compared to these solutions, Kilo emphasizes its open-source nature and flexibility in model selection. The company has secured $8 million in seed funding and reached a cooperation agreement with GitLab. Breitenother describes such tools as an "exoskeleton" for developer thinking, rather than a replacement for manual coding.