Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Sends prompts to OpenRouter using whichever models are currently free, with automatic fallback across multiple candidates. Retries on failures and returns structured JSON including the model used and response text.
Auto-discovery of free models means you don't need to manually track which OpenRouter models have a free tier or update your code when availability changes.
When to use it
- Running LLM queries at zero cost during prototyping
- Testing a prompt across multiple free models automatically
- Integrating AI responses into scripts without a paid API plan
- Getting a quick answer without choosing a specific model manually
- Logging which free models are available and responding on a given day
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Summarize the history of the Unix operating system
- 1Set OPENROUTER_API_KEY in the environment
- 2Run openrouter_free_chat.py with --prompt containing the user's question
- 3Script fetches current free model list and ranks candidates
- 4Script attempts models in order, retrying on failures
- 5Parse stdout JSON for selected_model and response fields
A text summary of Unix history, with a note that it was generated by e.g. meta-llama/llama-3-8b-instruct:free
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.