cn-llm

What This Skill Does

The OpenClaw CN-LLM skill serves as a unified gateway for accessing the top-tier large language models from the Chinese ecosystem. By abstracting the complexity of individual provider APIs, it offers a single, OpenAI-compatible endpoint that allows you to interact with models such as Qwen (Alibaba), DeepSeek, GLM, and Baichuan. This skill is designed for developers and power users who need high-performance Chinese language processing, advanced reasoning capabilities, and specialized coding assistance without managing multiple API keys or vendor-specific integration protocols. With the ability to switch between models like DeepSeek-R1 for logic-heavy tasks or Qwen3-Max for general intelligence, it acts as a centralized brain for your OpenClaw workflow.

Installation

To integrate this skill into your local environment, execute the following command in your terminal within the OpenClaw directory: clawhub install openclaw/skills/skills/chaimengphp/openclaw-aisa-cn-llm. Ensure that you have your AISA_API_KEY generated from the developer portal. Once installed, set your environment variable using export AISA_API_KEY="your-key-here". The skill will automatically detect this key for all subsequent API requests directed through the OpenClaw architecture.

Use Cases

This skill is ideal for several professional workflows. First, it excels in cross-lingual content generation and refinement where native Chinese context and nuance are required. Second, it is a powerhouse for software development, offering specialized coding models like DeepSeek-Coder to generate, refactor, or explain complex codebases. Third, it is highly effective for document analysis and long-form summarization when utilizing the 'Long' variants of the Qwen series, which are optimized for vast token windows. Finally, it provides an excellent testbed for model comparison, allowing users to benchmark the reasoning output of different architectures side-by-side.

Example Prompts

1. "Use DeepSeek-R1 to analyze the following logic error in my Python script and provide a fix: [Code Snippet]"
2. "Summarize this 50-page industry report into 5 key bullet points using Qwen3-Max in professional Chinese."
3. "Compare the output of DeepSeek-V3 and Qwen3-Max on the task of generating a creative marketing slogan for a new AI product."

Tips & Limitations

• Pricing Awareness: Costs are calculated per million tokens. Monitor your usage via the Aisa marketplace dashboard, as prices vary significantly between high-performance 'Max' models and 'Flash/Lite' efficiency variants.
• Rate Limiting: Always implement exponential backoff in your automation scripts to handle potential rate limits during heavy batch processing.
• Context Window: Be mindful that while models like Qwen-Long support massive text blocks, performance may degrade slightly in extreme edge cases; always chunk your input when feasible.
• Compatibility: Ensure your client library supports standard OpenAI-compliant configurations, as this is the baseline for all interaction with the gateway.