llm-db

Normalize provider and model metadata for deterministic catalog lookups in Elixir apps.

48|21|Updated Jan 25, 2025
One-click install
npx skills add https://github.com/agentjido/jido_run --skill llm-db-agentjido
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: llm-db
Source: https://github.com/agentjido/jido_run/tree/main/priv/skills/arrowcircle-jido-skills/skills/llm-db
Command: npx skills add https://github.com/agentjido/jido_run --skill llm-db-agentjido

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

LLM DB provides builder-oriented guidance to model provider catalogs, enabling Elixir apps to model, compare, and select models by capability and price, and to define clear boundaries with related components like req_llm, jido, and jido_ai.

Core Features & Use Cases

  • Build catalog-backed selection: create deterministic metadata lookups from provider data.
  • Add or refine provider metadata: preserve upstream IDs and map to local aliases when needed.
  • Turn docs into runnable examples: translate catalog docs into runnable Elixir workflows for lookups and decisions.
  • Review package boundaries: ensure catalog concerns stay in llm_db while route logic lives in the appropriate clients.

Quick Start

Use this skill to model provider metadata and build catalog lookups for model selection in Elixir.

Frequently Asked Questions about llm-db

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

FAQPage Schema
How do I build an LLM model catalog for deterministic lookups in Elixir?

You build an LLM model catalog by identifying, normalizing, and surfacing provider and model metadata to support deterministic catalog lookups across providers, modalities, and pricing constraints in Elixir apps.

What is the best way to compare LLM models by pricing and capability?

Comparing LLM models by pricing and capability requires applying catalog schemas and pricing metadata to your model selection workflow, enabling deterministic filtering across different providers and modalities.

How do I manage provider metadata and local aliases for LLM model selection?

Managing provider metadata for LLM model selection involves adding or refining provider data while preserving upstream IDs and mapping them to local aliases when needed for accurate catalog lookups.

Does this approach separate catalog concerns from route logic in Elixir applications?

Yes, package boundaries ensure catalog concerns stay in the LLM database while route logic lives in the appropriate clients, defining clear boundaries between catalog, req_llm, jido, and jido_ai components.

Can I translate LLM provider documentation into runnable Elixir workflows?

Yes, you can turn catalog documentation into runnable Elixir workflows that translate provider docs into executable examples for performing model lookups and automated selection decisions.