AgentGrade
Scan any website for AI agent readiness, payment protocols, and discovery endpoints
我该使用它吗
质量与安全性
A
发现（1）
- LOWTool 'scan_compact' description lacks action verb在 scan_compact 中
基于对工具定义和协议合规性的自动分析。
上下文开销
~293token 数（工具定义）
~419 B典型响应大小
对注意力的影响极小（占 128k 上下文窗口的 0.23%）
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": {
"mcp-server": {
"url": "https://agentgrade.com/mcp"
}
}
}远程端点
https://agentgrade.com/mcpstreamable-http它能做什么
工具清单
工具（4）
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪scan_url(url)
Scan a URL for agent capabilities and payment protocols. Returns full results including x402, MPP, L402, discovery, bazaar, MCP, plugins, OpenAPI, and more.
输入模式
{
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"description": "URL to scan (http or https)"
}
},
"required": [
"url"
]
}⚪scan_compact(url)
Compact scan returning only rails, capabilities, and a numeric score. Optimized for agent decision-making.
输入模式
{
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"description": "URL to scan"
}
},
"required": [
"url"
]
}🟢get_history(url, limit)
Get scan history for a URL. Requires database to be configured.
输入模式
{
"type": "object",
"properties": {
"url": {
"type": "string",
"description": "Filter by URL (optional)"
},
"limit": {
"type": "number",
"description": "Max results (default 20, max 100)"
}
}
}⚪validate_x402_json(content)
Validate x402.json content against expected schema. Returns errors, warnings, and suggestions. No network requests — pure validation.
输入模式
{
"type": "object",
"properties": {
"content": {
"type": "object",
"description": "The x402.json content to validate"
}
},
"required": [
"content"
]
}社区
证据
最近观测
已验证未记录版本4 个工具
已验证未记录版本4 个工具