Network
从实站爬取或粘贴的页面清单生成 llms.txt（llmstxt.org v2 格式：H1、引用块摘要、H2 文件列表），并按 RFC 9309 匹配算法审计 robots.txt 对 GPTBot、ClaudeBot、PerplexityBot、Bytespider 的策略——附 allow/block 预览与可直接粘贴的指令。
用三种语言从你的代码中调用此工具。
curl -X POST 'http://127.0.0.1:3003/zh/api/tools/llms-txt-and-ai-crawler-audit' \
-H 'Content-Type: application/json' \
-d '{"siteUrl":"https://example.com — homepage links, plus sitemap.xml fallback for JS-rendered sites (max 25 pages)","pagesInput":"\nhttps://example.com/docs/quickstart | Quickstart | Install the CLI and run your first job\nhttps://example.com/docs/api | API Reference | Endpoint catalog with auth examples\nhttps://example.com/blog/cli-tips | CLI Tips | Twenty shortcuts for daily use\nhttps://example.com/blog/release-2026 | Release 2026 | What changed this quarter","robotsInput":"User-agent: GPTBot\nDisallow: /\n\nUser-agent: ClaudeBot\nDisallow:\n\nUser-agent: *\nDisallow: /private/","siteName":"Example Platform Docs","summary":"Example Platform lets teams ship scheduled data jobs without infrastructure.","extraNotes":"All endpoints require an API key; the quickstart covers sandbox setup.","grouping":"auto"}'以 JSON 形式 POST 提交输入参数。文件类型参数需先单独上传。
POST http://127.0.0.1:3003/zh/api/tools/llms-txt-and-ai-crawler-audit| 参数名 | 类型 | 必填 | 说明 |
|---|---|---|---|
| siteUrl | text | 否 | — |
| pagesInput | textarea | 否 | — |
| robotsInput | textarea | 否 | — |
| siteName | text | 否 | — |
| summary | text | 否 | — |
| extraNotes | textarea | 否 | — |
| grouping | select | 是 | — |
HTML 结果
{
"result": "<div>Processed HTML content</div>",
"error": "Error message (optional)",
"message": "Notification message (optional)",
"metadata": {
"key": "value"
}
}将此工具加入你的 Model Context Protocol 服务，让 AI 智能体可以列出并调用它。
将以下内容加入你的 MCP 客户端配置：
{
"mcpServers": {
"elysiatools-llms-txt-and-ai-crawler-audit": {
"name": "llms-txt-and-ai-crawler-audit",
"description": "从实站爬取或粘贴的页面清单生成 llms.txt（llmstxt.org v2 格式：H1、引用块摘要、H2 文件列表），并按 RFC 9309 匹配算法审计 robots.txt 对 GPTBot、ClaudeBot、PerplexityBot、Bytespider 的策略——附 allow/block 预览与可直接粘贴的指令。",
"baseUrl": "http://127.0.0.1:3003/mcp/sse?toolId=llms-txt-and-ai-crawler-audit",
"command": "",
"args": [],
"env": {},
"isActive": true,
"type": "sse"
}
}
}连接到 SSE 端点后，列出已开放的工具：
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/list"
}通过工具 id 调用，参数由其参数表构建：
{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/call",
"params": {
"name": "llms-txt-and-ai-crawler-audit",
"arguments": {
"siteUrl": "https://example.com — homepage links, plus sitemap.xml fallback for JS-rendered sites (max 25 pages)",
"pagesInput": "\nhttps://example.com/docs/quickstart | Quickstart | Install the CLI and run your first job\nhttps://example.com/docs/api | API Reference | Endpoint catalog with auth examples\nhttps://example.com/blog/cli-tips | CLI Tips | Twenty shortcuts for daily use\nhttps://example.com/blog/release-2026 | Release 2026 | What changed this quarter",
"robotsInput": "User-agent: GPTBot\nDisallow: /\n\nUser-agent: ClaudeBot\nDisallow:\n\nUser-agent: *\nDisallow: /private/",
"siteName": "Example Platform Docs",
"summary": "Example Platform lets teams ship scheduled data jobs without infrastructure.",
"extraNotes": "All endpoints require an API key; the quickstart covers sandbox setup.",
"grouping": "auto"
}
}
}有问题或反馈？请联系 [email protected]