要点摘要

2026年的SEO基础是用robots.txt说明抓取规则，用sitemap.xml提交规范URL。两者不能互相冲突。

2026 robots.txt 与 sitemap.xml SEO 设置指南

robots.txt syntax: User-agent, Disallow, Allow, Sitemap

Use robots.txt at the root of the host, for example https://millionscode.com/robots.txt. The practical 2026 baseline is simple: open the parts that should be crawled, block low-value operational paths, and declare the sitemap with a full URL. A safe starting point is:

User-agent: *
Allow: /

Sitemap: https://millionscode.com/sitemap.xml

For a production site, treat Disallow carefully. A single Disallow: / can stop crawling of the whole host. That is useful on staging and dangerous on a live site. If admin pages, carts, search results, or temporary filters should not be crawled, block those paths only:

User-agent: *
Disallow: /admin/
Disallow: /search
Disallow: /cart
Allow: /blog/
Allow: /tools/

Sitemap: https://millionscode.com/sitemap.xml

robots.txt is not a security layer. It is a crawler instruction file. Sensitive data must be protected by authentication, and pages that must disappear from search need the right removal method rather than only a crawl block.

sitemap.xml structure

A sitemap lists canonical URLs that deserve discovery. Keep it clean:

<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
  <url>
    <loc>https://millionscode.com/tools/meta-checker</loc>
    <lastmod>2026-06-03</lastmod>
  </url>
  <url>
    <loc>https://millionscode.com/blog/post-mp90fbw1</loc>
    <lastmod>2026-06-03</lastmod>
  </url>
</urlset>

Do not mix canonical and non-canonical versions. Do not submit URLs blocked by robots.txt. Use /tools/meta-checker for metadata checks, /blog/post-mp90fbw1 for sitemap generation planning, /blog/api-mpj0hwex for indexing API workflow, and /blog/post-mpkfy95s for Search Console checks.

Submission: Search Console and Naver

In Google Search Console, verify the property and submit sitemap.xml or sitemap_index.xml in the Sitemaps report. Then check status, discovered URL count, fetch errors, and whether blocked URLs are accidentally present. In Naver Search Advisor, verify ownership, run robots.txt diagnosis, submit the sitemap, and watch collection requests separately. For Korean search traffic, Naver diagnostics should not be skipped.

Common mistakes

The first mistake is leaving a staging rule on production. The second is listing blocked URLs in the sitemap. The third is mixing www, non-www, http, and https versions. The fourth is changing lastmod every day without real content changes. The fifth is using robots.txt as an index removal tool. The reliable 2026 pattern is open crawl paths, submit only canonical URLs, and use console tools for diagnosis.

实务提示

For small and mid-size websites, the most useful habit is consistency. The pages in sitemap.xml should also be reachable through internal links. New posts, key tools, category pages, and evergreen guides should reinforce one another instead of living as isolated URLs. 每次发布路由、canonical、语言路径或自动XML生成逻辑后，都应重新检查这两个文件。

FAQ

robots.txt 会直接阻止索引吗？

不会。它主要控制抓取。删除或隐藏页面还需要 noindex、认证、删除工具或正确状态码。

Sitemap 行放在哪里？

使用完整 URL，通常放在 robots.txt 末尾。

Allow 一定优先吗？

通常更具体的路径优先，发布前要测试。

要放入所有 URL 吗？

只放规范且可索引的重要 URL。

提交后会马上收录吗？

不会，只是帮助发现和诊断。

Naver 也要提交吗？

面向韩国流量时建议提交并诊断。