voices

Query multiple LLMs in parallel via an OpenAI-compatible chat API.

6|1|Updated May 30, 2026
One-click install
npx skills add https://github.com/Wondermonger-daydreaming/claude-skills-library --skill voices
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: voices
Source: https://github.com/Wondermonger-daydreaming/claude-skills-library/tree/main/skills/voices
Command: npx skills add https://github.com/Wondermonger-daydreaming/claude-skills-library --skill voices

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Query multiple LLMs in parallel to surface diverse perspectives and cross-check insights.

Core Features & Use Cases

  • Cross-model dialogue: send a prompt to several hosts and compare their responses side by side.
  • Council diagnostics: quickly identify consensus vs. disagreement across architectures to inform decisions.
  • Open-ended testing: test ideas across model families (OpenAI, Anthropic, DeepSeek, etc.) using a single client.

Quick Start

Run the voices client with a set of models and a prompt to see concurrent responses.

Frequently Asked Questions about voices

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

FAQPage Schema
How do I query multiple LLMs in parallel to compare model responses?▼

Query multiple LLMs in parallel by sending a prompt to several model hosts concurrently, which surfaces diverse perspectives and cross-checks insights side by side. This concurrent request handling quickly identifies consensus or disagreement across architectures.

What is a council-style LLM comparison and when do I need it?▼

A council-style LLM comparison queries multiple model architectures simultaneously to identify consensus versus disagreement. You need this cross-model dialogue when making decisions that benefit from diverse AI perspectives and cross-checked insights.

Can I use a single API client to test prompts across OpenAI and Anthropic model families?▼

Yes, you can test ideas across model families like OpenAI, Anthropic, and DeepSeek using a single vendor-agnostic client. It satisfies an OpenAI-compatible chat API workflow to manage these cross-model dialogues.

How do I manage API keys for concurrent requests to different LLM hosts?▼

Manage API keys for concurrent requests through environment-based API key management. This setup securely handles authentication for multiple hosts when running cross-model dialogues and council-style comparisons.

Does this parallel LLM querying approach support saving the chat exchanges?▼

Yes, this parallel LLM querying approach supports optional saving of exchanges. While handling concurrent requests to surface diverse perspectives, you can save the cross-model dialogues for later analysis.