Which AI providers are natively supported in the registry?

The skill includes specific maintenance patterns for OpenAI, Anthropic (Claude), Google (Gemini), and xAI (Grok).

Does this skill support LiteLLM integration?

Yes, it provides logic to check the LiteLLM pricing database first, reducing the need for manual updates for over 500+ standard models.

What files does this skill help manage?

It specifically focuses on maintaining massgen/backend/capabilities.py for model features and release dates, and massgen/token_manager/token_manager.py for pricing and token limits.

How do I verify my changes after updating the registry?

The skill provides specific pytest commands to validate capabilities and token management, along with commands to test config generation.

When should I update the hardcoded PROVIDER_PRICING?

You should only update it if a model is missing from LiteLLM, if the LiteLLM data is outdated, or if you are configuring a custom internal model.

Model Registry Maintainer

Name: Model Registry Maintainer
Author: Microck

byMicrock

•

데이터 과학 및 ML

Maintains and updates model registries, pricing, and capabilities for AI providers within the MassGen ecosystem.

소개

The Model Registry Maintainer skill provides comprehensive guidance for managing the MassGen backend, specifically focusing on the capabilities and token management registries. It streamlines the process of adding new AI models, updating context windows, and adjusting pricing data across multiple providers like OpenAI, Anthropic, and Google. By leveraging integration patterns for LiteLLM and OpenRouter, this skill ensures that your AI implementation remains synchronized with the latest industry releases and cost structures while maintaining backend validation and documentation accuracy.

주요 기능

Step-by-step workflows for adding new models to capabilities.py
Automated validation using pytest and documentation generation scripts
Pricing and token limit management in token_manager.py
81 GitHub stars
Standardized release date and case-sensitive naming convention enforcement
Integration with LiteLLM and OpenRouter for real-time data fetching

사용 사례

Correcting outdated pricing or context window limits for legacy model versions
Adding support for custom or internal organization-specific AI models
Updating backend configurations when a new GPT, Claude, or Gemini model is released

소개

주요 기능

Step-by-step workflows for adding new models to capabilities.py
Automated validation using pytest and documentation generation scripts
Pricing and token limit management in token_manager.py
81 GitHub stars
Standardized release date and case-sensitive naming convention enforcement
Integration with LiteLLM and OpenRouter for real-time data fetching

사용 사례

Correcting outdated pricing or context window limits for legacy model versions
Adding support for custom or internal organization-specific AI models
Updating backend configurations when a new GPT, Claude, or Gemini model is released