api-flow-analyzer
Cloudflare Workers MCP server: api-flow-analyzer
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
~605token 数（工具定义）
~1.0 KB典型响应大小
对注意力的影响极小（占 128k 上下文窗口的 0.47%）
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": {
"api-flow-analyzer": {
"url": "https://api.lazy-mac.com/api-flow-analyzer/mcp"
}
}
}远程端点
https://api.lazy-mac.com/api-flow-analyzer/mcpstreamable-http它能做什么
工具清单
工具（6）
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪track_call(method, endpoint, duration_ms, response_size_bytes, status_code, ...)
Record and track an individual API call with metadata
输入模式
{
"type": "object",
"properties": {
"method": {
"type": "string",
"description": "HTTP method (GET, POST, etc.)"
},
"endpoint": {
"type": "string",
"description": "API endpoint URL"
},
"duration_ms": {
"type": "number",
"description": "Call duration in milliseconds"
},
"response_size_bytes": {
"type": "number",
"description": "Response payload size"
},
"status_code": {
"type": "number",
"description": "HTTP status code"
},
"service_name": {
"type": "string",
"description": "Source service identifier"
}
},
"required": [
"method",
"endpoint",
"duration_ms",
"status_code"
]
}🟢analyze_flow(calls, flow_id)
Analyze complete API call flow and identify patterns
输入模式
{
"type": "object",
"properties": {
"calls": {
"type": "array",
"description": "Array of API calls with metadata",
"items": {
"type": "object",
"properties": {
"method": {
"type": "string"
},
"endpoint": {
"type": "string"
},
"duration_ms": {
"type": "number"
},
"timestamp": {
"type": "number"
}
}
}
},
"flow_id": {
"type": "string",
"description": "Unique request flow identifier"
}
},
"required": [
"calls",
"flow_id"
]
}⚪detect_redundancy(calls, threshold_pct)
Identify duplicate or redundant API calls in a flow
输入模式
{
"type": "object",
"properties": {
"calls": {
"type": "array",
"description": "Array of API calls to analyze",
"items": {
"type": "object"
}
},
"threshold_pct": {
"type": "number",
"description": "Similarity threshold percentage (0-100)",
"default": 80
}
},
"required": [
"calls"
]
}⚪suggest_batching(calls, batch_size_limit)
Recommend batching opportunities for parallel API calls
输入模式
{
"type": "object",
"properties": {
"calls": {
"type": "array",
"description": "Array of API calls to optimize",
"items": {
"type": "object"
}
},
"batch_size_limit": {
"type": "number",
"description": "Max items per batch",
"default": 50
}
},
"required": [
"calls"
]
}🟢calculate_savings(current_calls, optimized_calls, cost_per_call_usd, avg_latency_ms)
Calculate estimated cost and latency savings from optimizations
输入模式
{
"type": "object",
"properties": {
"current_calls": {
"type": "number",
"description": "Current number of API calls"
},
"optimized_calls": {
"type": "number",
"description": "Optimized number of calls"
},
"cost_per_call_usd": {
"type": "number",
"description": "Cost per API call in USD",
"default": 0.0001
},
"avg_latency_ms": {
"type": "number",
"description": "Average latency per call in ms"
}
},
"required": [
"current_calls",
"optimized_calls"
]
}🟢get_dependency_graph(calls)
Generate dependency graph of API calls showing call sequence and relationships
输入模式
{
"type": "object",
"properties": {
"calls": {
"type": "array",
"description": "Array of API calls with dependencies",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string"
},
"endpoint": {
"type": "string"
},
"depends_on": {
"type": "array",
"items": {
"type": "string"
}
}
}
}
}
},
"required": [
"calls"
]
}社区
证据
最近观测
已验证未记录版本6 个工具
已验证未记录版本6 个工具