routing-services-courier
Pay-per-call ($0.01 USDC) model-routing for AI coding agents: which LLM to call, cost vs. quality.
我该使用它吗
质量与安全性
A
基于对工具定义和协议合规性的自动分析。
上下文开销
~151token 数(工具定义)
~1.4 KB典型响应大小
对注意力的影响极小(占 128k 上下文窗口的 0.12%)
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中:
{
"mcpServers": {
"routing-services-courier": {
"url": "https://routing-courier-mcp-production.up.railway.app/mcp"
}
}
}远程端点
https://routing-courier-mcp-production.up.railway.app/mcpstreamable-http它能做什么
工具清单
工具(1)
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪recommend_model_for_task(task_description, required_context_tokens, require_tool_calling)
Recommend which LLM to call for one specific task, balancing cost against a validated quality benchmark. Requires x402 payment ($0.01 USDC on Base).
输入模式
{
"type": "object",
"properties": {
"task_description": {
"title": "Task Description",
"type": "string"
},
"required_context_tokens": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"default": null,
"title": "Required Context Tokens"
},
"require_tool_calling": {
"default": false,
"title": "Require Tool Calling",
"type": "boolean"
}
},
"required": [
"task_description"
],
"title": "recommend_model_for_taskArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "recommend_model_for_taskDictOutput"
}社区
证据
最近观测
已验证未记录版本1 个工具
已验证未记录版本1 个工具