Run Claude Code with Kimi K2, GLM-4.6, Qwen3-Coder, or DeepSeek-v3.2 using a single command. Ideal for developers and operations teams automating code execution. Integrates with npm and supports Claude agents.
git clone https://github.com/kaichen/kimicc.gitKIMICC is a lightweight Node.js npm package that configures Claude Code to use alternative AI models including Kimi K2, GLM-4.6, Qwen3-Coder, and DeepSeek-v3.2 with a single command. It sets up environment variables automatically, eliminating manual configuration of Auth Tokens and Base URLs. The tool supports multiple service providers simultaneously through profile management, allowing developers to run concurrent requests and switch between models. Users run `npx kimicc` to initialize and select their preferred model provider, with persistent configuration stored for subsequent launches. This addresses accessibility and cost challenges by leveraging Anthropic-compatible interfaces from domestic and international vendors at a fraction of Claude's subscription cost.
[{"step":"Install kimicc globally via npm: `npm install -g kimicc` or use the standalone binary from the official repository.","tip":"Ensure you have Node.js v18+ installed for npm integration. Add `--global` flag if you want system-wide access."},{"step":"Prepare your code file with the correct extension (.js, .py, .ts, etc.) and ensure all dependencies are listed in package.json if using npm.","tip":"For non-JS projects, use `--no-npm` flag to skip dependency checks. Add `--env .env` to load environment variables."},{"step":"Execute the code using the kimicc command with your preferred model: `kimicc --model Qwen3-Coder /path/to/script.py`","tip":"Use `--watch` flag to automatically re-run when files change. Add `--timeout 30` for long-running processes."},{"step":"Review the output for execution results, performance metrics, and suggestions. Use `--json` flag for machine-readable output.","tip":"For debugging, add `--verbose` to see detailed model interaction logs. Pipe output to a file with `> output.log`"},{"step":"Integrate with CI/CD pipelines by adding kimicc as a step in your workflow files (GitHub Actions, GitLab CI, etc.).","tip":"Cache the model downloads between runs by setting `KIMICC_CACHE_DIR` environment variable to a persistent location."}]
Cost-effective code execution for developers in regions with Claude access restrictions
Running concurrent AI code tasks using multiple model providers simultaneously
Switching between Kimi, GLM, Qwen, and DeepSeek models without manual environment setup
Operations teams automating code execution with alternative LLM backends
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/kaichen/kimiccCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Run the following code using Kimi K2 with the command: `kimicc [OPTIONS] [FILE_PATH]` Code to execute: ```[LANGUAGE] [CODE] ``` [OPTIONS] (optional): Add flags like `--model Qwen3-Coder` or `--npm` if needed. Replace [LANGUAGE] and [CODE] with your actual code snippet.
```bash
$ kimicc --model GLM-4.6 --npm /path/to/script.js
Executing with Kimi K2 (GLM-4.6) and npm support...
✅ Code execution started
📦 Installing dependencies from package.json... (12 packages)
✨ Dependencies resolved in 2.1s
🚀 Running script.js with Node.js v20.12.2...
Output:
{
"status": "success",
"result": {
"processed_records": 150,
"errors": 0,
"warnings": 2,
"performance": {
"execution_time": "420ms",
"memory_usage": "18.4MB"
}
}
}
✅ Execution completed in 3.8s
📊 Performance metrics:
- Model: GLM-4.6 (context window: 128K tokens)
- npm packages: axios, lodash, moment (latest versions)
- Exit code: 0
💡 Suggestions:
1. Optimize the `processBatch` function to reduce memory usage by 15%
2. Consider caching the `fetchConfig` calls to improve performance
3. Add input validation for the `userId` parameterAI assistant built for thoughtful, nuanced conversation
Get more done every day with Microsoft Teams – powered by AI
Automate security compliance and monitor real-time security posture seamlessly.
Automate your spreadsheet tasks with AI power
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan