geo-llmstxt

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

llms.txt Standard Analysis and Generation Skill

llms.txt标准分析与生成技能

Purpose

用途

This skill handles everything related to the
llms.txt
standard -- an emerging convention (proposed by Jeremy Howard in September 2024, gaining adoption through 2025-2026) that allows websites to provide structured guidance to AI systems about their content, structure, and key information. It is analogous to
robots.txt
(which tells crawlers what NOT to access) but instead tells AI systems what IS most useful to understand about the site.
本技能处理与
llms.txt
标准相关的所有事宜——这是一项新兴约定(由Jeremy Howard于2024年9月提出,在2025-2026年期间逐渐被采用),允许网站向AI系统提供关于其内容、结构和关键信息的结构化指导。它类似于
robots.txt
(告诉爬虫哪些内容不能访问),但它是告诉AI系统网站中哪些内容最值得了解。

Why llms.txt Matters

为什么llms.txt很重要

AI language models face a fundamental challenge when processing websites: they must determine which pages are most important, what the site is about, and how content is organized -- typically by crawling many pages and inferring structure.
llms.txt
solves this by providing an explicit, machine-readable (and human-readable) summary.
Benefits of having a well-crafted llms.txt:
  1. Faster AI comprehension: AI systems can understand your site's purpose and structure from a single file rather than crawling dozens of pages.
  2. Controlled narrative: You choose which pages and facts AI systems see first, shaping how they represent your brand.
  3. Higher citation accuracy: AI systems that consult llms.txt can cite the correct, authoritative page for each topic.
  4. Reduced misrepresentation: Key facts (pricing, features, locations) are stated explicitly, reducing AI hallucination about your business.
  5. Early adopter advantage: As of early 2026, fewer than 5% of websites have an llms.txt file, making it a differentiator.

AI语言模型在处理网站时面临一个根本性挑战:它们必须确定哪些页面最重要、网站是关于什么的,以及内容是如何组织的——通常需要爬取大量页面并推断结构。
llms.txt
通过提供一个明确的、机器可读(同时也人类可读)的摘要来解决这个问题。
精心编写的llms.txt的优势:
  1. 更快的AI理解速度: AI系统无需爬取数十个页面,只需通过一个文件就能理解网站的用途和结构。
  2. 可控的叙事: 你可以选择AI系统首先看到哪些页面和事实,从而塑造品牌的呈现方式。
  3. 更高的引用准确性: 参考llms.txt的AI系统可以为每个主题引用正确的权威页面。
  4. 减少错误表述: 关键事实(定价、功能、地点)被明确说明,减少AI对业务的幻觉错误。
  5. 早期采用者优势: 截至2026年初,只有不到5%的网站拥有llms.txt文件,这使其成为一个差异化优势。

The llms.txt Specification

llms.txt规范

File Location

文件位置

The file MUST be located at the root of the domain:
https://example.com/llms.txt
文件必须位于域名的根目录下:
https://example.com/llms.txt

Format Specification

格式规范

The file uses Markdown formatting with specific conventions:
markdown
undefined
文件使用Markdown格式,并遵循特定约定:
markdown
undefined

[Site Name]

[网站名称]

[One-sentence description of what the site/business does. Keep under 200 characters.]
[一句话描述网站/业务的内容,控制在200字符以内。]

Docs

文档

  • Page Title: Concise description of what this page covers and why it matters.
  • Another Page: Description of content.

Optional

可选

undefined
undefined

Detailed Format Rules

详细格式规则

1. Title (Required)
markdown
undefined
1. 标题(必填)
markdown
undefined

Site Name

网站名称

- Must be the first line of the file.
- Should be the official business/site name.
- Use the H1 heading format (single `#`).

**2. Description (Required)**
```markdown
> Brief description of the site/business
  • Must appear immediately after the title.
  • Use Markdown blockquote format (
    >
    ).
  • Keep under 200 characters.
  • Should clearly state what the business does and who it serves.
  • Avoid marketing fluff -- be factual and specific.
3. Main Sections (Required -- at least one)
Use H2 headings (
##
) to organize pages by category. Common section names:
Section NamePurposeExample Content
## Docs
Primary documentation or key pagesProduct pages, service descriptions, core content
## Optional
Secondary pages worth knowing aboutBlog posts, supplementary resources
## API
API documentationAPI reference, authentication guides
## Blog
Blog or news contentRecent/popular articles
## Products
Product catalogProduct pages, pricing
## Services
Service offeringsService descriptions, process pages
## About
Company informationAbout page, team, mission
## Resources
Educational/reference contentGuides, tutorials, whitepapers
## Legal
Legal documentsTerms of service, privacy policy
## Contact
Contact informationContact page, support channels
4. Page Entries (Required)
Each entry follows the format:
markdown
- [Page Title](URL): Description of page content
Rules for page entries:
  • Title: Use the actual page title or a clear descriptive title.
  • URL: Must be a full, absolute URL (not relative paths).
  • Description: 10-30 words describing what the page covers. Be specific about the information available.
  • Order: List pages in order of importance within each section.
  • Limit: Include 10-30 page entries total. Prioritize your most authoritative and useful pages.
5. Key Facts Section (Recommended)
markdown
undefined
- 必须是文件的第一行。
- 应使用官方的业务/网站名称。
- 使用H1标题格式(单个`#`)。

**2. 描述(必填)**
```markdown
> 网站/业务的简要描述
  • 必须紧跟在标题之后。
  • 使用Markdown块引用格式(
    >
    )。
  • 控制在200字符以内。
  • 应清晰说明业务内容和服务对象。
  • 避免营销空话——保持事实性和具体性。
3. 主要章节(必填——至少一个)
使用H2标题(
##
)按类别组织页面。常见章节名称:
章节名称用途示例内容
## 文档
主要文档或关键页面产品页面、服务描述、核心内容
## 可选
值得了解的次要页面博客文章、补充资源
## API
API文档API参考、认证指南
## 博客
博客或新闻内容近期/热门文章
## 产品
产品目录产品页面、定价
## 服务
服务项目服务描述、流程页面
## 关于我们
公司信息关于页面、团队、使命
## 资源
教育/参考内容指南、教程、白皮书
## 法律
法律文件服务条款、隐私政策
## 联系我们
联系信息联系页面、支持渠道
4. 页面条目(必填)
每个条目遵循以下格式:
markdown
- [页面标题](URL): 页面内容描述
页面条目规则:
  • 标题: 使用实际页面标题或清晰的描述性标题。
  • URL: 必须是完整的绝对URL(不能是相对路径)。
  • 描述: 10-30个单词,描述页面涵盖的内容。具体说明可用的信息。
  • 顺序: 在每个章节内按重要性排序页面。
  • 数量限制: 总共包含10-30个页面条目。优先选择最权威和最有用的页面。
5. 关键事实章节(推荐)
markdown
undefined

Key Facts

关键事实

  • Founded in [year] by [founder(s)]
  • Headquarters: [City, Country]
  • customers/users in [Y] countries
  • Key products: [Product A], [Product B], [Product C]
  • Industry: [Industry classification]

This section provides quick reference data that AI systems frequently need to answer user queries about your business.

**6. Contact Section (Recommended)**

```markdown
  • 成立于[年份],创始人[创始人姓名]
  • 总部:[城市,国家]
  • 在[Y]个国家拥有[X]个客户/用户
  • 核心产品:[产品A]、[产品B]、[产品C]
  • 行业:[行业分类]

本章节提供AI系统在回答用户关于业务的查询时经常需要的快速参考数据。

**6. 联系我们章节(推荐)**

```markdown

Contact

联系我们

  • Website: https://example.com
  • Email: hello@example.com
  • Support: support@example.com
  • Phone: +1-555-123-4567
  • Address: 123 Main St, City, State, ZIP, Country

---
  • 网站:https://example.com
  • 邮箱:hello@example.com
  • 支持:support@example.com
  • 电话:+1-555-123-4567
  • 地址:123 Main St, City, State, ZIP, Country

---

llms-full.txt (Extended Version)

llms-full.txt(扩展版本)

In addition to
llms.txt
, sites can provide
/llms-full.txt
-- an extended version with more detail.
Differences from llms.txt:
Featurellms.txtllms-full.txt
LengthConcise (50-150 lines)Comprehensive (150-500+ lines)
Page entries10-30 key pages30-100+ pages
Descriptions10-30 words per entry30-100 words per entry, may include key facts from each page
AudienceQuick AI comprehensionDeep AI analysis
Sections3-6 sections8-15 sections
Key factsBusiness-level factsPage-level facts and data points
Both files can coexist. AI systems check for
llms.txt
first, then may optionally load
llms-full.txt
for deeper understanding.

除了
llms.txt
,网站还可以提供
/llms-full.txt
——包含更多细节的扩展版本。
与llms.txt的区别:
特性llms.txtllms-full.txt
长度简洁(50-150行)全面(150-500+行)
页面条目10-30个关键页面30-100+个页面
描述每个条目10-30个单词每个条目30-100个单词,可能包含每个页面的关键事实
受众快速AI理解深度AI分析
章节3-6个章节8-15个章节
关键事实业务层面的事实页面层面的事实和数据点
两个文件可以共存。AI系统会先检查
llms.txt
,然后可以选择加载
llms-full.txt
以获得更深入的理解。

Analysis Mode

分析模式

When checking an existing llms.txt file:
当检查现有llms.txt文件时:

Step 1: Fetch the File

步骤1:获取文件

  1. Use WebFetch to retrieve
    [domain]/llms.txt
    .
  2. Also check for
    [domain]/llms-full.txt
    .
  3. Record HTTP status code:
    • 200: File exists -- proceed to validation.
    • 404: File does not exist -- recommend generation.
    • 403: File exists but is blocked -- flag as misconfiguration.
    • 301/302: Redirect -- follow and note the redirect.
  1. 使用WebFetch获取
    [域名]/llms.txt
  2. 同时检查
    [域名]/llms-full.txt
  3. 记录HTTP状态码:
    • 200: 文件存在——继续验证。
    • 404: 文件不存在——建议生成。
    • 403: 文件存在但被阻止——标记为配置错误。
    • 301/302: 重定向——跟随并重定向记录。

Step 2: Validate Format

步骤2:验证格式

Check each structural element:
ElementCheckSeverity if Missing
H1 TitlePresent, matches business nameCritical
Blockquote descriptionPresent, under 200 chars, factualHigh
At least one H2 sectionPresentCritical
Page entries with URLsAt least 5 entries presentHigh
URLs are absoluteAll URLs use full https:// pathsHigh
URLs are validAll URLs return 200 statusMedium
Descriptions presentEvery entry has a description after the colonMedium
Key Facts sectionPresent with business informationMedium
Contact sectionPresent with at least emailLow
Reasonable length30-200 linesLow
No broken MarkdownProper formatting throughoutMedium
检查每个结构元素:
元素检查内容缺失的严重程度
H1标题存在,与业务名称匹配严重
块引用描述存在,200字符以内,符合事实
至少一个H2章节存在严重
带URL的页面条目至少存在5个条目
URL为绝对路径所有URL使用完整的https://路径
URL有效所有URL返回200状态码
条目描述存在每个条目在冒号后都有描述
关键事实章节存在并包含业务信息
联系我们章节存在并至少包含邮箱
长度合理30-200行
Markdown无错误整个文件格式正确

Step 3: Assess Content Quality

步骤3:评估内容质量

Rate the llms.txt on these dimensions:
Completeness (0-100):
  • Does it cover all major site sections visible in the navigation?
  • Are the most important/highest-traffic pages included?
  • Is the Key Facts section present with accurate business data?
  • Does it include recent/updated content?
Accuracy (0-100):
  • Do descriptions accurately reflect page content?
  • Are URLs valid and pointing to the correct pages?
  • Are Key Facts verifiable and current?
  • Is the business description accurate?
Usefulness (0-100):
  • Would an AI system understand the site's purpose from this file alone?
  • Are descriptions specific enough to differentiate pages?
  • Are the most citation-worthy pages highlighted?
  • Is the organization logical and intuitive?
Overall llms.txt Score = (Completeness * 0.40) + (Accuracy * 0.35) + (Usefulness * 0.25)
从以下维度为llms.txt评分:
完整性(0-100):
  • 是否涵盖了导航中可见的所有主要网站章节?
  • 是否包含了最重要/流量最高的页面?
  • 是否存在包含准确业务数据的关键事实章节?
  • 是否包含了最新/更新的内容?
准确性(0-100):
  • 描述是否准确反映页面内容?
  • URL是否有效并指向正确页面?
  • 关键事实是否可验证且是最新的?
  • 业务描述是否准确?
实用性(0-100):
  • AI系统仅通过此文件能否理解网站的用途?
  • 描述是否足够具体以区分不同页面?
  • 是否突出了最值得引用的页面?
  • 组织逻辑是否直观合理?
llms.txt总体得分 = (完整性 * 0.40) + (准确性 * 0.35) + (实用性 * 0.25)

Step 4: Compare Against Site Content

步骤4:与网站内容对比

  1. Crawl the site's main navigation and sitemap.
  2. Identify important pages NOT listed in llms.txt.
  3. Check if any listed URLs are broken or redirected.
  4. Verify that the business description matches current homepage messaging.
  5. Flag stale entries (pages that have been significantly updated since the llms.txt was written).

  1. 爬取网站的主导航和站点地图。
  2. 识别llms.txt中未列出的重要页面。
  3. 检查列出的URL是否有损坏或重定向。
  4. 验证业务描述是否与当前首页信息一致。
  5. 标记过时条目(自llms.txt编写以来页面内容已大幅更新)。

Generation Mode

生成模式

When creating a new llms.txt file from scratch:
当从头创建新的llms.txt文件时:

Step 1: Site Discovery

步骤1:网站发现

  1. Fetch the homepage and extract:
    • Site name (from
      <title>
      ,
      <meta property="og:site_name">
      , or H1)
    • Business description (from meta description or hero section)
    • Main navigation links
    • Footer links
  2. Fetch
    /sitemap.xml
    to discover all public pages.
  3. Identify the site's primary business type (SaaS, E-commerce, Local, Publisher, Agency).
  1. 获取首页并提取:
    • 网站名称(来自
      <title>
      <meta property="og:site_name">
      或H1)
    • 业务描述(来自元描述或英雄区域)
    • 主导航链接
    • 页脚链接
  2. 获取
    /sitemap.xml
    以发现所有公开页面。
  3. 确定网站的主要业务类型(SaaS、电商、本地商家、出版商、代理机构)。

Step 2: Page Prioritization

步骤2:页面优先级排序

Categorize all discovered pages and select the most important ones:
Always Include:
  • Homepage
  • About / Company page
  • Pricing page (if exists)
  • Primary product/service pages (top 3-5)
  • Contact page
  • Documentation landing page (if exists)
Include if High Quality:
  • Top blog posts (by apparent importance, recency, or comprehensiveness)
  • Case studies or customer stories
  • Key resource/guide pages
  • FAQ page
  • Careers page (for large companies)
Skip:
  • Thin category/tag pages
  • Pagination pages
  • Login/signup pages
  • Legal boilerplate (unless specifically relevant)
  • Duplicate or near-duplicate content
  • Pages with minimal unique content
对所有发现的页面进行分类,并选择最重要的页面:
必须包含:
  • 首页
  • 关于我们/公司页面
  • 定价页面(如果存在)
  • 核心产品/服务页面(前3-5个)
  • 联系我们页面
  • 文档首页(如果存在)
如果质量高则包含:
  • 热门博客文章(根据重要性、时效性或全面性判断)
  • 案例研究或客户故事
  • 关键资源/指南页面
  • FAQ页面
  • 招聘页面(适用于大公司)
跳过:
  • 内容单薄的分类/标签页面
  • 分页页面
  • 登录/注册页面
  • 法律模板(除非特别相关)
  • 重复或近乎重复的内容
  • 独特内容极少的页面

Step 3: Write Descriptions

步骤3:编写描述

For each selected page:
  1. Fetch the page content using WebFetch.
  2. Read the H1, meta description, and first 2-3 paragraphs.
  3. Write a description that:
    • Is 10-30 words long
    • States what information is on the page
    • Mentions specific topics, data, or features covered
    • Avoids marketing language ("best," "leading," "revolutionary")
    • Uses factual, informative language
Good description examples:
  • Explains the three pricing tiers (Free, Pro, Enterprise) with feature comparison and annual/monthly costs.
  • Details the company's founding in 2018, team of 45 employees, and office locations in Austin and London.
  • Covers integration setup for Slack, Salesforce, and HubSpot with step-by-step guides and API endpoints.
Bad description examples:
  • Our amazing pricing page!
    (marketing language, no specifics)
  • Learn more about our company.
    (too vague)
  • Click here for details.
    (not descriptive)
对于每个选定的页面:
  1. 使用WebFetch获取页面内容。
  2. 阅读H1、元描述和前2-3段。
  3. 编写描述,要求:
    • 长度为10-30个单词
    • 说明页面上的信息内容
    • 提及涵盖的特定主题、数据或功能
    • 避免营销语言(如“最佳”、“领先”、“革命性”)
    • 使用事实性、信息性语言
好的描述示例:
  • 解释三个定价层级(免费、专业、企业版),包含功能对比和年/月付费用。
  • 详细介绍公司2018年成立、45名员工以及奥斯汀和伦敦的办公地点。
  • 涵盖Slack、Salesforce和HubSpot的集成设置,包含分步指南和API端点。
差的描述示例:
  • 我们超棒的定价页面!
    (营销语言,无具体信息)
  • 了解更多关于我们公司的信息。
    (过于模糊)
  • 点击此处查看详情。
    (无描述性)

Step 4: Compile Key Facts

步骤4:整理关键事实

Gather key business facts from the site:
  • Year founded
  • Founder name(s)
  • Headquarters location
  • Number of employees (if public)
  • Number of customers/users (if public)
  • Key products or services (list top 3-5)
  • Industry classification
  • Notable clients or partnerships (if public)
  • Key differentiators (what makes this business unique)
  • Recent milestones or achievements (last 12 months)
从网站收集关键业务事实:
  • 成立年份
  • 创始人姓名
  • 总部位置
  • 员工人数(如果公开)
  • 客户/用户数量(如果公开)
  • 核心产品或服务(列出前3-5个)
  • 行业分类
  • 知名客户或合作伙伴(如果公开)
  • 核心差异化优势(业务的独特之处)
  • 近期里程碑或成就(过去12个月)

Step 5: Assemble the File

步骤5:组装文件

Construct the llms.txt following this template:
markdown
undefined
按照以下模板构建llms.txt:
markdown
undefined

[Site Name]

[网站名称]

[One clear sentence: what the business does, who it serves, and its primary value proposition. Under 200 characters.]
[清晰的一句话:业务内容、服务对象和核心价值主张,控制在200字符以内。]

Docs

文档

Products

产品

  • Product A: Core features, target users, and pricing model for Product A.
  • Product B: What Product B does and how it differs from Product A.
  • 产品A: 产品A的核心功能、目标用户和定价模式。
  • 产品B: 产品B的功能及其与产品A的区别。

Resources

资源

  • Guide Title: Comprehensive guide covering [topic] with [X] sections and practical examples.
  • Blog Post: Analysis of [topic] with original data from [source].
  • 指南标题: 全面指南,涵盖[主题],包含[X]个章节和实用示例。
  • 博客文章: 分析[主题],包含来自[来源]的原始数据。

Key Facts

关键事实

  • Founded in [year] by [name(s)]
  • Headquartered in [City, Country]
  • [Specific metric: e.g., "Serves 10,000+ businesses in 40 countries"]
  • [Key differentiator: e.g., "Only platform offering real-time X and Y integration"]
  • Industry: [Classification]
  • 成立于[年份],创始人[姓名]
  • 总部位于[城市,国家]
  • [具体指标:例如“为40个国家的10,000+企业提供服务”]
  • [核心差异化优势:例如“唯一提供实时X和Y集成的平台”]
  • 行业:[分类]

Contact

联系我们

undefined
undefined

Step 6: Validate the Generated File

步骤6:验证生成的文件

Before outputting:
  1. Verify all URLs are reachable (200 status).
  2. Confirm total entry count is between 10-30.
  3. Check that no description exceeds 50 words.
  4. Verify the overall file length is 50-150 lines.
  5. Ensure Markdown formatting is clean and consistent.

输出前:
  1. 验证所有URL均可访问(200状态码)。
  2. 确认条目总数在10-30之间。
  3. 检查所有描述不超过50个单词。
  4. 验证文件总长度在50-150行之间。
  5. 确保Markdown格式清晰一致。

Output Format

输出格式

For Analysis Mode

分析模式

Generate
GEO-LLMSTXT-ANALYSIS.md
:
markdown
undefined
生成
GEO-LLMSTXT-ANALYSIS.md
markdown
undefined

llms.txt Analysis: [Domain]

llms.txt分析:[域名]

Analysis Date: [Date] llms.txt Status: [Found at URL / Not Found / Error] llms-full.txt Status: [Found / Not Found]

分析日期: [日期] llms.txt状态: [在URL找到/未找到/错误] llms-full.txt状态: [找到/未找到]

Overall llms.txt Score: [X]/100

llms.txt总体得分:[X]/100

DimensionScore
Completeness[X]/100
Accuracy[X]/100
Usefulness[X]/100

维度得分
完整性[X]/100
准确性[X]/100
实用性[X]/100

Format Validation

格式验证

ElementStatusNotes
H1 Title[Pass/Fail][Notes]
Description blockquote[Pass/Fail][Notes]
H2 Sections[Pass/Fail][X sections found]
Page entries[Pass/Fail][X entries found]
URL validity[Pass/Fail][X broken URLs]
Entry descriptions[Pass/Fail][X missing descriptions]
Key Facts[Pass/Fail][Notes]
Contact section[Pass/Fail][Notes]

元素状态备注
H1标题[通过/未通过][备注]
块引用描述[通过/未通过][备注]
H2章节[通过/未通过][找到X个章节]
页面条目[通过/未通过][找到X个条目]
URL有效性[通过/未通过][X个损坏的URL]
条目描述[通过/未通过][X个缺失的描述]
关键事实[通过/未通过][备注]
联系我们[通过/未通过][备注]

Missing Pages

缺失页面

These important pages were found on the site but not in llms.txt:
  1. Page Title -- [Why it should be included]
  2. Page Title -- [Why it should be included]
这些网站上的重要页面未在llms.txt中列出:
  1. 页面标题 -- [应包含的原因]
  2. 页面标题 -- [应包含的原因]

Improvement Recommendations

改进建议

  1. [Specific recommendation]
  2. [Specific recommendation]
  3. [Specific recommendation]
  1. [具体建议]
  2. [具体建议]
  3. [具体建议]

Suggested Updated llms.txt

建议更新后的llms.txt

[Complete rewritten llms.txt file if significant improvements are needed]
undefined
[如果需要重大改进,提供完整重写的llms.txt文件]
undefined

For Generation Mode

生成模式

Output the complete
llms.txt
file content, ready to be saved to the site's root directory. Also output a brief
GEO-LLMSTXT-GENERATION.md
report explaining:
  • How many pages were discovered and how many were selected
  • The prioritization rationale
  • Any pages that were borderline (might add later)
  • Recommended update frequency (e.g., monthly for active blogs, quarterly for stable sites)

输出完整的
llms.txt
文件内容,可直接保存到网站根目录。同时输出简短的
GEO-LLMSTXT-GENERATION.md
报告,说明:
  • 发现的页面数量和选择的页面数量
  • 优先级排序依据
  • 哪些页面属于边缘情况(以后可能添加)
  • 建议的更新频率(例如,活跃博客每月更新,稳定网站每季度更新)

Best Practices Reference

最佳实践参考

  1. Update regularly. If your site publishes weekly blog posts, update llms.txt monthly. If your product changes quarterly, update after each release.
  2. Lead with your strongest content. The first entries in each section should be your most authoritative, comprehensive pages.
  3. Be specific in descriptions. "Comprehensive 3,000-word guide to React Server Components with code examples" is far more useful than "React guide."
  4. Include your differentiators. If your site has unique data, original research, or exclusive features, highlight these in descriptions and Key Facts.
  5. Keep it concise. The llms.txt should be scannable in under 60 seconds. Save detail for llms-full.txt.
  6. Use absolute URLs. Always include the full
    https://
    URL, never relative paths.
  7. Test after deployment. After uploading, verify the file is accessible at
    https://yourdomain.com/llms.txt
    with no redirects.
  8. Coordinate with robots.txt. Ensure pages listed in llms.txt are not blocked in robots.txt for AI crawlers.
  9. Mirror your site structure. Section names in llms.txt should roughly correspond to your main navigation categories.
  10. Avoid sensitive pages. Do not include internal tools, admin panels, or pages with sensitive information.
  1. 定期更新。如果网站每周发布博客文章,每月更新llms.txt。如果产品每季度更新,每次发布后更新。
  2. 优先展示最优质内容。每个章节的第一个条目应为最权威、最全面的页面。
  3. 描述要具体。“包含代码示例的3000字React Server Components全面指南”比“React指南”有用得多。
  4. 突出差异化优势。如果网站有独特数据、原创研究或独家功能,在描述和关键事实中突出这些内容。
  5. 保持简洁。llms.txt应能在60秒内快速浏览。详细内容放在llms-full.txt中。
  6. 使用绝对URL。始终包含完整的
    https://
    URL,绝不要使用相对路径。
  7. 部署后测试。上传后,验证文件可在
    https://yourdomain.com/llms.txt
    访问,无重定向。
  8. 与robots.txt协调。确保llms.txt中列出的页面未被robots.txt阻止AI爬虫访问。
  9. 镜像网站结构。llms.txt中的章节名称应大致对应主导航类别。
  10. 避免敏感页面。不要包含内部工具、管理面板或包含敏感信息的页面。