llm-sparring

jbo-tech/llm-sparring

★ 0 stars Python AI/LLM Updated today

An MCP server that orchestrates sparring sessions between LLMs.

View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "llm-sparring": {
      "command": "uvx",
      "args": [
        "llm-sparring"
      ]
    }
  }
}

Or install with pip: pip install llm-sparring

README Excerpt

**Make your LLMs disagree.** An MCP server that orchestrates sparring sessions between LLMs. Query multiple models, have them challenge each other, sharpen your ideas through friction. - **Multi-provider**: OpenAI, Anthropic, Google, OpenRouter, Moonshot (Kimi), z.ai (GLM), Ollama, custom endpoints - **Reasoning-model aware**: Parser unifié qui gère `reasoning_content`, `reasoning` (OpenRouter), `<think>` inline (phi-4, QwQ), `max_completion_tokens` (GPT-5 / o-series), `finish_reason` exposé au

Tools (17)

anthropicask_modelcustomdeepseekestimate_costget_usagegooglegroqmistralmoonshotollamaopenaiopenroutersession_idtogetherxaizai

llm-sparring

Quick Install

README Excerpt

Tools (17)

Related AI/LLM Servers