freeride

What This Skill Does

The FreeRide skill for OpenClaw is an intelligent management system designed to optimize your AI model usage by leveraging the free tier offerings on OpenRouter. Manually switching between models to avoid costs or hit rate limits can be tedious; FreeRide automates this process by ranking available models based on context length, capabilities, and reliability. It continuously monitors your configuration, updating your openclaw.json file to ensure you are always using the most effective primary model while maintaining a robust list of fallback providers. By setting a primary model for consistent performance and utilizing a prioritized stack of backups—including OpenRouter’s native routing—the skill ensures that your workflow remains uninterrupted even if a specific provider encounters capacity issues.

Installation

Installing FreeRide is a straightforward process managed via the ClawHub CLI. To get started, open your terminal and execute the following command:

npx clawhub@latest install freeride

Once installed, you must ensure your OpenRouter API key is available to the environment. Obtain your key from openrouter.ai/keys and add it to your shell configuration (e.g., .bashrc or .zshrc) as an environment variable: export OPENROUTER_API_KEY="sk-or-v1-...". After setting the key, you can run freeride auto to let the skill analyze your environment and apply the optimal default configuration.

Use Cases

FreeRide is designed for power users who want to maximize their AI output without incurring costs. It is ideal for developers who utilize multiple LLMs for different tasks and need a reliable, cost-effective infrastructure that handles transient model failures gracefully. It also serves those who frequently switch between specialized models—such as coding-centric models versus general-purpose chat models—by allowing quick swaps while maintaining an automated fallback layer.

Example Prompts

1. "I'm tired of hitting rate limits with my current AI setup, can you switch my configuration to use free models from OpenRouter?"
2. "Please refresh my model list and set the primary to a high-performance coding model while keeping 5 fallback options."
3. "Can you optimize my OpenClaw setup to use only free models to save on API costs?"

Tips & Limitations

To get the most out of FreeRide, ensure you keep your OpenRouter API key updated and periodically use the freeride refresh command to sync with the latest model availability. Note that while this skill excels at managing configuration, it depends on OpenRouter's platform stability. If you encounter errors, verify your file permissions for ~/.openclaw/openclaw.json, as the skill requires write access to update your agent settings. Remember that 'free' models may have different rate-limit profiles compared to paid variants, so keeping a healthy number of fallbacks is recommended for heavy usage.