Claude Code + MiniMax-M2.5 Integration for Openclaw

A professional configuration guide to empower the Claude Code CLI with the high-performance MiniMax-M2.5 language model for advanced AI programming.

l1-m1ng
v1.0.0
Mar 5, 2026
0
973
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install claude-code-minimax

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install claude-code-minimax using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Claude Code + MiniMax-M2.5 Integration?

This skill provides a comprehensive bridge between Anthropic's Claude Code and the MiniMax-M2.5 model, enabling developers to use a localized, high-speed LLM for complex software engineering tasks. By utilizing Openclaw Skills, developers can bypass default API constraints and leverage MiniMax-M2.5 for reading and writing code, executing terminal commands, and managing project-wide refactoring.

Whether you are looking for region-specific model performance or cost-effective alternatives to standard backends, this integration ensures that your AI coding agent remains highly capable within a professional CLI environment.

Claude Code + MiniMax-M2.5 Integration Use Cases

  • Generating complex Python scripts for data processing and automation.
  • Real-time debugging by providing error logs directly to the AI agent.
  • Performing project-wide code quality audits and architectural refactoring.
  • Writing unit tests for critical source files automatically.
  • Initializing new project templates for React, TypeScript, and other modern frameworks.

How Claude Code + MiniMax-M2.5 Integration Works

  1. The user navigates to their project directory and initializes the AI agent.
  2. The system requests folder-level permissions to ensure secure file manipulation.
  3. All natural language prompts are routed through the configured MiniMax-M2.5 endpoint using the Anthropic-compatible API layer.
  4. The agent analyzes the local codebase and proposes specific modifications or terminal commands.
  5. The developer reviews the proposed changes and confirms execution with a single keystroke.

Claude Code + MiniMax-M2.5 Integration Setup

There are three primary ways to configure this integration as part of your Openclaw Skills workflow:

Install the visual configuration tool to manage API keys easily:

# macOS / Linux
brew tap farion1231/ccswitch
brew install --cask cc-switch
  1. Launch cc-switch and add a new MiniMax configuration.
  2. Set the model name to MiniMax-M2.5.
  3. Enable the configuration and ensure ~/.claude.json has {"hasCompletedOnboarding": true}.

Option 2: Manual Configuration

Edit ~/.claude/settings.json and include the following environment mapping:

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.minimaxi.com/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_MINIMAX_API_KEY",
    "ANTHROPIC_MODEL": "MiniMax-M2.5"
  }
}

Claude Code + MiniMax-M2.5 Integration Data Schema & Taxonomy

The skill manages its operational state and model routing through specific local files and environment variables:

File Path Purpose
~/.claude/settings.json Stores environment variables like ANTHROPIC_BASE_URL and model mappings.
~/.claude.json Flag for onboarding status to bypass initial CLI prompts.
Environment Variables Overrides file settings for API_TIMEOUT_MS and security traffic filters.

Claude Code + MiniMax-M2.5 Integration Advanced Features

  • Universal model aliasing: Redirects all Claude sub-models (Sonnet, Opus, Haiku) to MiniMax-M2.5 for consistent behavior.
  • Visual API switching: Rapidly toggle between different AI providers using cc-switch integration.
  • VS Code Extension support: Seamlessly inject MiniMax environment variables into the VS Code integrated terminal.
  • Verbose debugging mode: Execute claude --verbose to inspect raw API traffic and model reasoning chains.
  • High-latency optimization: Configurable API_TIMEOUT_MS up to 3000s for massive code generation tasks.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*