AI AGENTS
Codive
by Codive
Overview
Open-source terminal AI coding agent for understanding codebases, editing files, running commands, searching code, and working with existing editors across multiple LLM providers.
Details
Codive is described by its official site and documentation as an open-source terminal AI coding agent that understands codebases, edits files, runs commands, supports multiple LLM providers, and works with existing editors. Its usage documentation covers interactive and one-shot modes and says it can read files, create or modify files, run shell commands, and search codebases. Installation documentation lists Cargo, Homebrew, script, Windows, GitHub Releases, and source-build paths.
When to Use
Use Codive when you want a terminal-based AI coding agent that can operate inside an existing development workflow. Use it for codebase tasks that require reading searching creating or modifying files. Use it when you want an agent that can run shell commands and work with multiple LLM providers.
Getting Started
- Read the introduction: https://codive.dev/docs/introduction/
- Install Codive using one of the documented paths: Cargo
- Homebrew
- install script
- Windows
- GitHub Releases
- or source build: https://codive.dev/docs/installation/
- Follow the quick-start workflow: install Codive
- provide an Anthropic or OpenAI API key
- run `codive` in a project
- then use interactive or one-shot prompts: https://codive.dev/docs/quickstart/
- Review basic usage for interactive mode
- one-shot mode
- file operations
- shell commands
- and codebase search: https://codive.dev/docs/usage/basic/
Key Features
- •Terminal AI coding agent
- •Codebase understanding
- •Read
- •create
- •and modify files
- •Run shell commands
- •Search codebases
- •Interactive and one-shot modes
- •Multiple LLM provider support
- •Works with existing editors
Capabilities
- •codebase understanding
- •file reading
- •file creation and modification
- •shell command execution
- •codebase search
- •interactive coding assistance
- •one-shot task execution
- •multi-provider LLM support
Last updated Jun 12, 2026