Install AI Tools

B2C Commerce tools, documentation, and skills for your assistant.

Claude

Install the plugin Recommended

bash
claude plugin marketplace add SalesforceCommerceCloud/b2c-developer-tooling
claude plugin install b2c-dx-mcp@b2c-developer-tooling --scope project

Start a new Claude Code session in your project. Use --scope user instead for all projects.

Manual MCP setup

From your project directory:

bash
claude mcp add --transport stdio --scope project b2c-dx-mcp -- npx -y @salesforce/b2c-dx-mcp@latest

Start a new session. Use --scope user instead for all projects. See Claude Code MCP setup.

Claude Desktop setup

Codex

Install the plugin Recommended

bash
codex plugin marketplace add SalesforceCommerceCloud/b2c-developer-tooling
codex plugin add b2c-dx-mcp@b2c-developer-tooling

Start a new Codex session in your project. This setup also works with the Codex IDE extension and the ChatGPT Work desktop app.

Manual MCP setup
bash
codex mcp add b2c-dx-mcp -- npx -y @salesforce/b2c-dx-mcp@latest

Or add this to ~/.codex/config.toml (or $CODEX_HOME/config.toml if customized):

toml
[mcp_servers.b2c-dx-mcp]
command = "npx"
args = ["-y", "@salesforce/b2c-dx-mcp@latest"]

Start a new session. See Codex MCP configuration.

ChatGPT online setup

VS Code

Install in VS Code

Install the plugin Recommended

  1. Open the Command Palette (Cmd/Ctrl+Shift+P) and run Chat: Install Plugin from Source.
  2. Enter SalesforceCommerceCloud/b2c-developer-tooling.
  3. Select b2c-dx-mcp and follow the installation prompts.
  4. Start a new chat in GitHub Copilot.
Manual MCP setup

Add this to .vscode/mcp.json in your workspace:

json
{
  "servers": {
    "b2c-dx-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@salesforce/b2c-dx-mcp@latest"]
    }
  }
}

See VS Code MCP setup.

Copilot CLI setup

Cursor

Install in Cursor

Reload the MCP server in Cursor after installation.

Manual MCP setup

Add this to .cursor/mcp.json in your project:

json
{
  "mcpServers": {
    "b2c-dx-mcp": {
      "command": "npx",
      "args": ["-y", "@salesforce/b2c-dx-mcp@latest"]
    }
  }
}

For all projects, use ~/.cursor/mcp.json instead.

See Cursor's MCP documentation.

OpenCode

Add this to opencode.json in your project:

json
{
  "mcp": {
    "b2c-dx-mcp": {
      "type": "local",
      "command": ["npx", "-y", "@salesforce/b2c-dx-mcp@latest"],
      "enabled": true
    }
  }
}

Restart OpenCode. For all projects, use ~/.config/opencode/opencode.json. See OpenCode MCP setup.

Gemini

From your project directory, run:

bash
gemini mcp add --scope project b2c-dx-mcp -- npx -y @salesforce/b2c-dx-mcp@latest

Start a new Gemini CLI session. Use --scope user instead for all projects. See Gemini CLI MCP setup.

No separate skills plugins needed.

Other clients and manual setup →
Agentic B2C Developer Toolkit

Appearance

View as Markdown
View as Markdown

Script Debugger ​

The B2C Commerce Script Debugger lets you set breakpoints, step through code, and inspect variables in server-side scripts — SFRA controllers, hooks, jobs, custom SCAPI endpoints, or any dw/* cartridge code — running live on an instance. You can drive it from the VS Code extension, another IDE, the CLI, or an AI agent.

Requirements ​

The debugger needs Basic auth credentials — a Business Manager username and either the account password or a WebDAV File Access and UX Studio access key (used as the password). OAuth/client credentials are not sufficient.

The debugger uses the same resolved credentials as the rest of the CLI (flags, SFCC_* environment variables, or dw.json). See the Authentication Guide for access key setup and Configuration for how credentials are resolved.

Choosing an interface ​

Use caseInterfaceReference
Debug from VS Code (recommended)Salesforce B2C Commerce VS Code ExtensionVS Code Extension
Debug from another IDE (JetBrains)b2c debug (DAP debug adapter)Debug Commands
Debug with your AI assistantB2C MCPDebug with your assistant

The VS Code extension is the recommended interface for interactive debugging — it provides the full graphical debugger (breakpoints, log points, watch expressions, step controls), just like any other Node project. The CLI's DAP debug adapter (b2c debug) also offers a headless terminal mode for scripting; see Debug Commands for details.

They all share the same workflow: connect a session, set breakpoints (by local file path, cartridge-prefixed path, or server path), trigger the code on the instance, then inspect the halted thread.

Rare PIG-only troubleshooting: If a breakpoint is never hit after confirming the request exercises the expected code and the source mapping is correct, the request may be reaching a different app server on a multi-app-server Production Instance Group. In that case, use Copy Debugger Session ID (dwsid) in VS Code or the MCP session's session_cookie, then send the triggering request with Cookie: dwsid=<value> (or sfdc_dwsid: <value> for a headless request). Sandboxes are single-app-server and never need this.

Debug with your assistant ​

With the B2C MCP, your assistant can investigate live cartridge execution: set a breakpoint, inspect variables and call stacks, and step through the code to test an explanation. Use it for an unexpected controller response, a failing hook, or a custom job step whose inputs differ from what you expected.

Share the relevant source and a way to reproduce the issue in your sandbox. Your assistant can connect the observed values to the code and Script API documentation, then explain the finding before you decide whether to change anything.

Example prompt

This custom job step skips products I expect it to update. Inspect its inputs and filtering logic with the debugger while I run it in my sandbox. Explain which condition excludes a product, then resume and disconnect. Don't change the code.

Breakpoints pause requests or jobs. Use a sandbox; evaluating expressions can change application state. The MCP uses the same debugger credentials. See MCP debugger access for permissions and precautions.

See Also ​