Route prompts to multiple LLM providers with configurable routing rules.
npx skills add https://github.com/ZealousEar/claude-skills --skill llm-zealousear
Please help me install this Agent Skill. Skill: llm Source: https://github.com/ZealousEar/claude-skills/tree/main/claude-code/llm Command: npx skills add https://github.com/ZealousEar/claude-skills --skill llm-zealousear
Route prompts to multiple providers from a single interface, enabling cross-model comparisons and guardrail enforcement. The system supports CLI-first routing for Claude Code, Codex, and Kimi, and API-based routing for Google, OpenRouter, and Anthropic models, with auto-discovery of new models. It relies on a local registry, per-model prompting overrides, and routing rules to deterministically select providers and gracefully fall back when needed.
Prompt the router to send a query to a chosen model (e.g., gpt-5.3-codex) and return its response.
High-intent search queries and answers about installing and using this skill.
You can route prompts to multiple LLM providers from a single interface by using configurable routing rules, per-model prompting overrides, and a local registry to deterministically select providers and gracefully fall back when needed.
Multi-provider LLM routing handles fallback by using deterministic routing rules that gracefully shift execution to alternative models when the primary provider fails, ensuring task completion.
Yes, you can use CLI-first routing for Claude Code, Codex, and Kimi alongside API-based routing for Google, OpenRouter, and Anthropic models, with auto-discovery of new models.
You enforce guardrails and apply per-model prompting overrides across different models by relying on deterministic task execution via scripts and structured routing rules within the unified interface.
The best way to compare responses across different LLM models is to use a unified routing interface that sends identical prompts to multiple providers, enabling direct cross-model comparisons.
Yes, multi-provider LLM routing supports auto-discovery of new models, automatically registering them in the local registry for immediate use with existing routing rules and overrides.
SternrasslerDesign Go backend service layers with constructor injection and Redis caching.
tadod12Extract structured data from LLM responses with Pydantic validation and automatic retries.
tblakex01Benchmark the same prompt across Claude, GPT, and Gemini for latency, tokens, and cost.
netlifyRoute AI model calls through Netlify AI Gateway with automatic environment variable injection.
