Use Corti Models in your coding agent

Connect OpenCode, Pi, Crush, or ForgeCode to Corti Models with the Corti CLI.

Connect your terminal AI coding agent to Corti Models. Install a supported agent, run the Corti CLI, and start coding.

This is the coding agent path for agentic coding and personal usage. If you want to replace OpenAI or Anthropic as the LLM provider in your application code, see Replace your LLM provider instead.

US projects: corti-s1 model IDs are served by Azure-hosted OpenAI models. See Regions.

Before you start

You’ll need:

• A terminal on macOS, Linux, or Windows
• Node.js 18+ installed (required to run the Corti CLI via npx)
• A Corti Console account (free to create)

The Corti CLI writes a corti provider entry into each selected agent’s config and manages a .env block with your credentials. Each config references CORTI_BEARER rather than embedding the secret itself.

1. Install a coding agent

   The Corti CLI supports the following agents. Install one (or more) using your preferred method.

   OpenCode

   Install

   curl -fsSL https://opencode.ai/install | bash
   

   Verify

   opencode --version
   

   Learn more at opencode.ai.

   OpenCode Desktop

   Download the desktop UI for macOS, Linux, or Windows.

   Download OpenCode Desktop

   Get the installer at opencode.ai/download.

   ForgeCode

   Install

   curl -fsSL https://forgecode.dev/cli | sh
   

   Verify

   forge --version
   

   Learn more at forgecode.dev.

   Crush

   Install

   curl -fsSL https://charm.land/install.sh | bash
   

   Verify

   crush --version
   

   Learn more at charm.land/crush.

   Pi

   Install

   curl -fsSL https://pi.dev/install.sh | sh
   

   Verify

   pi --version
   

   Learn more at pi.dev.

2. Run the setup wizard

   Run the interactive setup to authenticate and configure the Corti provider for your installed agents:

   npx @corti/cli models init
   

   npx requires Node.js 18+. Download it from nodejs.org or install via a version manager like fnm or nvm.

   The wizard opens by asking how you want to sign in:

   ◆ How do you want to sign in?
   │  ▸ ● Log in with a browser
   │    ○ Paste an API key
   │    ○ Enter client credentials
   └ ↑↓ move · enter select
   

   Press Enter to accept the default, Log in with a browser. The CLI shows a one-time code and opens the Corti Console. Approve the sign-in there, and the terminal continues on its own.

   If the browser can’t open automatically, for example over SSH, the CLI prints a URL for you to open instead.

   Paste an API key instead

   1. In the Console, open Corti Models and generate your personal API key.
   2. In the CLI, choose Paste an API key and paste it.

   The value is masked on entry and stored as CORTI_BEARER.

   Client credentials still work as well, mainly for headless and CI runs.

   Once you’re signed in, the wizard:

   1. Verifies the connection and probes /models so you can pick a default model.
   2. Detects installed agents and pre-selects them (toggle with Space).
   3. Lets you choose scope: Global (~/.config), Project (current directory), or both.
   4. Shows a plan and asks you to confirm before writing anything.

   Press Enter to accept the defaults at each step.

   Where the CLI writes your config

   The CLI writes a corti provider entry into each selected agent’s config file, plus a managed .env block with your credentials:

   • ForgeCode — ~/.forge/.forge.toml
   • OpenCode — ~/.config/opencode/opencode.json (or .jsonc)
   • Crush — ~/.config/crush/crush.json
   • Pi — ~/.pi/agent/models.json
3. Load your credentials

   If you let the CLI update your shell startup file, open a new terminal and skip ahead.

   Otherwise, load the .env manually. Global scope writes it to ~/.env, project scope to .env where you ran the CLI.

   macOS / Linux

   bash / zsh

   set -a; . ~/.env; set +a
   

   Windows

   PowerShell

   Get-Content "$HOME\.env" | ForEach-Object {
       $line = $_.Trim()
       if ($line -and -not $line.StartsWith('#')) {
           $name, $value = $line -split '=', 2
           Set-Item -Path "Env:$($name.Trim())" -Value $value.Trim().Trim('"').Trim("'")
       }
   }
   
4. Start coding

   Launch your agent — you’re now running an AI coding assistant backed by Corti Models.

   OpenCode

   opencode
   

   ForgeCode

   forge
   

   Crush

   crush
   

   Select provider corti when prompted.

   Pi

   pi
   

   Select provider corti when prompted.

   Ask the agent a question and confirm it answers using the Corti model you selected.

Next steps

AI Coding Tools

Prompt Claude Code, Cursor, Codex, or Lovable with a Corti build skill.

API Reference

Browse the full Corti API with interactive examples.

JavaScript SDK

SDK reference, the foundation the CLI and skills build on.

Agentic Framework

Multi-expert orchestration, MCP servers, and the agent runtime.

AI-generated code should always be reviewed and tested before use in production. Human oversight ensures correctness, security, and compliance with applicable regulations.