add-model-descriptions

This skill automates the synchronization and documentation process for new LLM models within the chat-ui repository. It is designed for engineers and maintainers who need to keep their local or production deployment configuration up to date with the latest models available via the HuggingFace Inference Router. By fetching the model list from the API and cross-referencing it with existing entries in prod.yaml and dev.yaml, this tool identifies missing models, performs necessary research on architecture and capabilities, and formats descriptions to meet specific style guidelines.

• Automatically fetches available models from the HuggingFace router endpoint.

• Identifies discrepancies between the router and existing deployment configurations.

• Generates concise, professional model descriptions following architectural and capability-focused standards.

• Updates both prod.yaml and dev.yaml with new model entries seamlessly.

• Includes standardized commits to ensure version control integrity for configuration changes.

• Ensure the environment provides read/write access to the chat-ui configuration directory.

• Follow the defined descriptive style: 8-12 words, sentence fragments, and focusing on specific capabilities like MoE, coding, or vision-language tasks.

• Useful for maintaining up-to-date model metadata for chat interfaces, supporting efficient model discovery and user selection.

• Designed for use with SvelteKit-based chat deployments using OpenAI-compatible APIs.