mcpindex
The MCP server index that vets servers, not just lists them. Advisory screen before you install.
我该使用它吗
质量与安全性
发现（2）
- LOWTool 'recommend_mcp_for_task' description lacks action verb在 recommend_mcp_for_task 中
- LOWTool 'compare_servers' description lacks action verb在 compare_servers 中
基于对工具定义和协议合规性的自动分析。
上下文开销
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-mcpindex": {
"command": "npx",
"args": [
"mcp-server-mcpindex"
]
}
}
}可运行的软件包
0.4.0stdio远程端点
https://mcpindex.ai/api/mcpstreamable-http它能做什么
工具清单
工具（6）
⚪recommend_mcp_for_task(task)
Recommend the best MCP servers for a natural-language task. Returns top 3 ranked picks with reasoning, install commands, and quality scores. Use this when the user asks for the right MCP server for a task they want to do.
输入模式
{
"type": "object",
"properties": {
"task": {
"type": "string",
"description": "Natural-language description of the task, e.g. \"read PDFs and write to S3\"."
}
},
"required": [
"task"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢search_mcp_servers(query, category, limit)
Keyword + semantic search across every MCP server mcpindex indexes. Use when the user knows what tool category they want but not which server.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Search query."
},
"category": {
"description": "Optional category filter (e.g. database, browser, github).",
"type": "string"
},
"limit": {
"description": "Max results (default 10, max 50).",
"type": "integer",
"minimum": 1,
"maximum": 50
}
},
"required": [
"query"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_install_command(server_slug, client)
Get the exact install command for a given MCP server and client. Returns a JSON block ready to paste into the client config.
输入模式
{
"type": "object",
"properties": {
"server_slug": {
"type": "string",
"description": "Slug of the server (from search or recommend results)."
},
"client": {
"type": "string",
"enum": [
"claude-desktop",
"claude-code",
"cursor",
"gemini-cli",
"cline"
],
"description": "Target client."
}
},
"required": [
"server_slug",
"client"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}⚪compare_servers(slugs)
Side-by-side comparison of 2-5 MCP servers - quality scores, install paths, env vars.
输入模式
{
"type": "object",
"properties": {
"slugs": {
"minItems": 2,
"maxItems": 5,
"type": "array",
"items": {
"type": "string"
},
"description": "Server slugs to compare."
}
},
"required": [
"slugs"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢check_tool_trust(server_id, tool_name)
Pre-invocation advisory screen for a specific tool on an MCP server. Returns an advisory verdict object (directive ALLOW | DENY | REVIEW | UNVERIFIED, dimensions, freshness). At v1 the public screen produces REVIEW or UNVERIFIED only - ALLOW/DENY are reserved. Not the in-path gate (mcpindex-gate). Agents SHOULD treat UNVERIFIED as "human review required", never as ALLOW.
输入模式
{
"type": "object",
"properties": {
"server_id": {
"type": "string",
"description": "Registry slug from search_mcp_servers / recommend results, e.g. \"io-github-microsoft-playwright-mcp\" (NOT a short name like \"github\")."
},
"tool_name": {
"type": "string",
"description": "Tool name as exposed by the server (e.g. \"create_pull_request\")."
}
},
"required": [
"server_id",
"tool_name"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢assess_server(server_id)
Aggregated pre-flight trust assessment across all tools on an MCP server. Same verdict shape as check_tool_trust. Use for "is THIS server worth integrating?" decisions. v1 advisory; conformance monitored not enforced; verdicts may be UNVERIFIED if not yet probed.
输入模式
{
"type": "object",
"properties": {
"server_id": {
"type": "string",
"description": "Server slug to assess."
}
},
"required": [
"server_id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}社区
证据