skills.rest
Consiliency
ConsiliencyOFFICIAL
Explore all skills by this author →
WebsiteGitHub

model-discovery

Discover and classify AI model names from provider APIs.

8|1|Updated Jul 11, 2025
Search Tags:#caching#AI-models#model-discovery#new-model-detection#provider-apis#tier-classification
One-click install
npx skills add https://github.com/Consiliency/treesitter-chunker --skill model-discovery
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: model-discovery
Source: https://github.com/Consiliency/treesitter-chunker/tree/main/.ai-dev-kit/skills/model-discovery
Command: npx skills add https://github.com/Consiliency/treesitter-chunker --skill model-discovery

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python-dotenv, and includes scripts (resource) components.

What problem does it solve?

This Skill keeps AI model references current by automatically discovering model names and their tier classifications from provider APIs, ensuring accurate routing and model selection for downstream tasks.

Core Features & Use Cases

  • Real-time model discovery: Fetches model names from Anthropic, OpenAI, Gemini, and Ollama, capturing IDs, display names, and creation dates.
  • Tier classification: Classifies models into fast/default/heavy using naming patterns and config rules, flagging new models for review.
  • Use case: When configuring agents, rely on an up-to-date model list and automatic tier mapping for routing decisions.

Quick Start

  • Run: uv run python scripts/fetch_models.py --force
  • Check for new models: uv run python scripts/check_new_models.py --json
  • Auto-classify new models: uv run python scripts/check_new_models.py --auto

Frequently Asked Questions about model-discovery

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I keep AI model names up to date across multiple providers?▼

Model discovery automatically fetches current model names and IDs from Anthropic, OpenAI, Gemini, and Ollama APIs, then caches the results. Run `uv run python scripts/fetch_models.py --force` to refresh your model catalog and ensure downstream tasks use valid, current model references.

Can I automatically classify models into performance tiers?▼

Yes. Model discovery classifies models into fast, default, and heavy tiers using naming patterns and configuration rules. Use `uv run python scripts/check_new_models.py --auto` to automatically tier new models, or run without `--auto` to review classifications interactively before applying them.

What's the best way to detect when new models are released?▼

Run `uv run python scripts/check_new_models.py --json` to detect newly released models from provider APIs. The Skill flags new models for review, letting you identify when providers add models and decide whether to integrate them into your agent routing logic.

Does model discovery work with Ollama and commercial providers together?▼

Yes. Model discovery supports Anthropic, OpenAI, Gemini, and Ollama in a single workflow. Fetch model names from all providers simultaneously, apply consistent tier classification across them, and route agents to the appropriate model based on a unified, up-to-date catalog.

How do I integrate fresh model data into my agent configuration?▼

Model discovery outputs model names, display names, and tier classifications compatible with agent routing decisions. Cache fresh data automatically, then feed the output into downstream agent configuration or model selection scripts that rely on current provider model availability.

Related Skills

Explore More Software Engineering Skills→
ax-llmax-llm
2.9k

ax-python-ai

Create Python provider clients for axllm across OpenAI, Gemini, and Anthropic workflows.

Official
Intermediate
gabergergaberger
4

hex-inference

Configure inference providers and map local models to hex agent tiers.

Community
Advanced
ViperJuiceViperJuice
8

model-discovery

Fetch and classify AI provider models into fast, default, and heavy tiers.

Community
Advanced
in-the-loop-labsin-the-loop-labs
48

Update Provider Models

Update AI provider model configurations in pair-review source code files.

Official
Intermediate