Documentation
This documentation will help you get the most out of cecli.
- Installation — How to install and get started pair programming with cecli.
-
Usage — How to use cecli to pair program with AI and edit code in your local git repo.
- Tips — Tips for AI pair programming with cecli.
- In-chat commands — Control cecli with in-chat commands like /add, /model, etc.
- Chat modes — Using the code, architect, ask and help chat modes.
- Voice-to-code with cecli — Speak with cecli about your code!
- Images & web pages — Add images and web pages to the cecli coding chat.
- cecli in your IDE — cecli can watch your files and respond to AI comments you add in your favorite IDE or text editor.
- Notifications — cecli can notify you when it's waiting for your input.
- Specifying coding conventions — Tell cecli to follow your coding conventions when it works on your code.
- Copy/paste with web chat — cecli works with LLM web chat UIs
- Session Management — Session management utilities to save and load work across multiple application runs
- Linting and testing — Automatically fix linting and testing errors.
- Editing config & text files — Use cecli to edit configuration files, documentation, and other text-based formats.
- Connecting to LLMs — cecli can connect to most LLMs for AI pair programming.
-
Configuration — Information on all of cecli's settings and how to use them.
- Agent Mode — Agent Mode enables autonomous codebase exploration and modification using local tools.
- API Keys — Setting API keys for API providers.
- Options reference — Details about all of cecli's settings.
- YAML config file — How to configure cecli with a YAML config file.
- Security Configuration — Security configuration for controlling access to external resources and implementing security policies.
- Config with .env — Using a .env file to store LLM API keys for cecli.
- Custom Commands — Create and use custom commands to extend cecli's functionality.
- Custom System Prompts — Create and use custom system prompts to tailor the AI's behavior for specific use cases.
- Hooks — Create and use custom commands to extend cecli's functionality.
- Skills System — Extend AI capabilities with custom instructions, reference materials, scripts, and assets through the skills system.
- TUI Mode — TUI (Textual User Interface) Mode provides a modern, visually rich terminal interface for AI pair programming.
- Workspaces — Workspaces allow you to work across multiple related repositories simultaneously
- Editor configuration — How to configure a custom editor for cecli's /editor command
- Reasoning models — How to configure reasoning model settings from secondary providers.
- Model Control Protocol (MCP) — Configure Model Control Protocol (MCP) servers for enhanced AI capabilities.
- Advanced model settings — Configuring advanced settings for LLMs.
- Model Aliases — Assign convenient short names to models.
- Troubleshooting — How to troubleshoot problems with cecli and get help.
-
More info
- Git integration — cecli is tightly integrated with git.
- Supported languages — cecli supports pretty much all popular coding languages.
- Repository map — cecli uses a map of your git repository to provide code context to LLMs.
- Scripting cecli — You can script cecli via the command line or python.
- Infinite output — cecli can handle "infinite output" from models that support prefill.
- Edit formats — cecli uses various "edit formats" to let LLMs edit source files.
- Analytics — Opt-in, anonymous, no personal info.
- Privacy policy
- FAQ — Frequently asked questions about cecli.
-
LLM Leaderboards — Quantitative benchmarks of LLM code editing skill.
- Code editing leaderboard — Quantitative benchmark of basic LLM code editing skill.
- Refactoring leaderboard — Quantitative benchmark of LLM code refactoring skill.
- Scores by release date
- Contributing results