add-siliconflow-provider

What This Skill Does

This skill, add-siliconflow-provider, is designed to seamlessly integrate the SiliconFlow AI model inference platform into your OpenClaw environment. SiliconFlow is a prominent domestic platform offering a wide array of over 98 chat models, including popular series like Qwen, DeepSeek, Kimi, GLM, and MiniMax. A key advantage of SiliconFlow is its provision of multiple free models (e.g., Qwen3-8B, DeepSeek-R1-8B) and its competitive pricing for premium models, often at 30-50% of official rates. The integration utilizes the standard OpenAI openai-completions protocol, ensuring easy compatibility. This skill automates the entire process of adding SiliconFlow as a model source, including provider registration, model definition, alias configuration, and fallback linking. It is triggered by natural language requests indicating a desire to add or configure SiliconFlow models.

Installation

To install the add-siliconflow-provider skill, use the following command:

clawhub install openclaw/skills/skills/jooey/add-siliconflow-provider

This command will fetch the skill from the openclaw/skills repository. It's recommended to back up your existing OpenClaw configuration before proceeding:

cp ~/.openclaw/openclaw.json ~/.openclaw/openclaw.json.backup.$(date +%Y%m%d_%H%M%S)

The skill will then guide you through adding the SiliconFlow provider, which involves configuring essential parameters like the provider name (siliconflow), API protocol (openai-completions), Base URL (https://api.siliconflow.cn/v1), and authentication method (Bearer Token via API Key).

Use Cases

This skill is invaluable for users looking to:

• Expand Model Access: Gain access to a broad spectrum of Chinese AI models not readily available elsewhere.
• Reduce Costs: Leverage significantly cheaper model inference options provided by SiliconFlow, especially for paid models.
• Utilize Free Models: Take advantage of the numerous free-to-use models offered by SiliconFlow for testing and development.
• Streamline Configuration: Automate the complex process of setting up a new AI model provider with standard protocols.
• Enhance AI Capabilities: Integrate powerful models like Qwen, DeepSeek, and Kimi into your workflows for advanced natural language processing tasks.

Example Prompts

• "我想配置 SiliconFlow 作为模型源。"
• "请帮我添加 Kimi 和 Qwen3 的模型。"
• "为 OpenClaw 接入 SF 的免费模型。"

Tips & Limitations

• API Key is Crucial: Ensure you have a valid SiliconFlow API Key obtained from their console. The skill requires this for authentication. You can create one at https://cloud.siliconflow.cn/account/ak.
• Free Models: While free models like Qwen3-8B and DeepSeek-R1-8B are available, check SiliconFlow's terms for any usage limits. Paid models require sufficient balance in your SiliconFlow account.
• Model Aliases: The skill supports configuring recommended aliases for easier model selection (e.g., sf-qwen3-8b, sf-kimi). You can customize these if needed.
• OpenAI Compatibility: The openai-completions protocol ensures broad compatibility, but always verify the specific model endpoints and parameters if you encounter issues.
• Network Access: This skill requires network access to register the provider and potentially to test API keys with SiliconFlow's endpoints. Ensure your network allows outbound connections to api.siliconflow.cn.
• Configuration Changes: The skill modifies your openclaw.json configuration file. Always keep a backup, as demonstrated in the installation steps.