Recently, an open-source AI Agent CLI tool called DailiCode has attracted attention on GitHub. As a fork of the Gemini CLI, this tool offers powerful multi-model compatibility and rich features, providing developers with a new efficient and convenient command-line AI assistant experience.

 Cross-platform compatibility, breaking down model barriers

The biggest highlight of DailiCode is its wide range of model adaptation capabilities. It not only supports mainstream large language model (LLM) providers such as Gemini and OpenAI, but also can be compatible with any custom LLM API that follows the OpenAI API format. With simple environment variable configuration, developers can seamlessly switch between different models without changing their existing workflow.

Test data shows that the tool has completed multi-dimensional verification for 15 models, including Google Gemini-2.5-pro, OpenRouter Claude Sonnet4, Gpt-4.1, etc. Among them, six models, such as Google Gemini-2.5-pro and OpenRouter Grok-4, have met all seven indicators, including thinking process, multimodal ability, and complex task processing, demonstrating comprehensive performance advantages.

QQ20250725-092107.png

Feature aggregation, activating development efficiency

This command-line tool integrates multiple practical functions: it supports querying and editing large codebases using large context windows, generates applications from PDFs or sketches through multimodal capabilities, automates handling of Pull Request queries, complex rebases, and other operations, and can extend new features through MCP servers.

In practical applications, developers can use it to quickly generate Discord bots, summarize code changes, migrate codebase versions, and even achieve batch conversion of image formats, classification of PDF invoices, and other system interaction tasks. By introducing code via NPM, it can also fulfill customized needs such as generating sales data charts.

Lightweight deployment, lowering the usage threshold

DailiCode has a very low deployment threshold. Just install Node.js version 20 or higher, and you can quickly start it with the command "npx https://github.com/nearmetips/DailiCode", or complete local deployment through npm installation. The configurable options it provides, such as temperature parameters and maximum token numbers, allow developers to precisely control model output according to their needs.

As an open-source project developed based on easy llm cli and Gemini CLI, DailiCode provides developers with a new solution for connecting tools, understanding code, and accelerating workflows. Whether exploring new codebase architectures, handling GitHub Issues, or automating team collaboration processes, this tool shows great potential in reshaping development efficiency and will undoubtedly become an important force in the field of AI-assisted programming.