llms.txt: The Complete Setup Guide for South African Websites

What Is llms.txt?

Direct answer: llms.txt is a plain-text file placed at the root of your website — at yourdomain.co.za/llms.txt — that tells AI language model crawlers how to understand and index your site's content. Think of it as the AI-era version of robots.txt, except instead of telling crawlers what NOT to index, it tells them what your site is about and where your best content lives.

Current as of 31 May 2026: This article has been reviewed for the 2026 South African AI, SEO, and automation market. Pricing, platform capabilities, Google rich-result rules, and AI model features change quickly, so verify live vendor documentation before procurement. For privacy and data handling, use the Protection of Personal Information Act as the baseline; for search and structured-data implementation, use Google Search Central.

The format was proposed in 2024 and has since been adopted as a de facto standard by the GEO community. AI systems from OpenAI (ChatGPT), Anthropic (Claude), Perplexity, and others actively check for this file when crawling websites.

For South African businesses, implementing llms.txt is one of the highest-return GEO investments you can make. It takes under an hour to create, requires no developer skills, and can meaningfully improve how AI systems understand and represent your business.

Why llms.txt Matters

Without llms.txt, AI crawlers must infer your business identity, services, expertise, and content hierarchy from the raw HTML of your pages. This inference works reasonably well for large, well-structured sites. For most South African business websites, it produces an incomplete or distorted picture.

With llms.txt, you directly control the narrative:

• You tell AI crawlers exactly what your business does
• You highlight your most authoritative content pages
• You provide context that your website design may obscure (credentials, SA-specific expertise, POPIA compliance)
• You give crawlers a starting map of your site rather than requiring them to discover everything by crawling

The llms.txt Format

llms.txt uses a simple Markdown-like format. Here is the basic structure:

# [Your Business Name]

[One paragraph description of what your business does, who you serve, and where you operate.]

## [Section Name]
- [Description of page or section]: URL

## [Another Section]
- [Description]: URL

The file should be:

• Plain text (UTF-8 encoding)
• Accessible publicly at your-domain.co.za/llms.txt
• No authentication required
• Updated when your content significantly changes

A Real Example for a South African AI Consultancy

# Smart AI Solutions

Smart AI Solutions is a South African AI implementation consultancy specializing in process automation, AI chatbots, and data analytics for SMEs and enterprises. Founded in 2022 and headquartered in
Cape Town, South Africa. We serve clients across Johannesburg, Durban, Pretoria, and Cape Town.

## Services
- AI Process Automation for South African businesses: /services/process-automation
- AI Chatbots and Virtual Assistants: /services/ai-chatbots-assistants
- Data Analytics and Business Intelligence: /services/data-analytics-reporting
- AI Team Enablement and Training: /services/ai-team-enablement
- CRM and Sales Automation: /services/crm-sales-automation

## GEO Optimization Tools
- Free GEO Audit for South African websites: /tools/geo-audit

## Blog
- Practical AI and GEO guides for South African businesses: /blog

## About
- Company background, team credentials, and client work: /about

## Contact
- Get in touch for a free consultation: /contact

## Key Facts
- Location: Cape Town, South Africa
- Founded: 2022
- Specialization: AI implementation for South African SMEs and enterprises
- Languages: English, Afrikaans
- POPIA compliant

This example takes about 30 minutes to write and gives AI crawlers a complete, unambiguous picture of the business.

What to Include in Your llms.txt

Required Elements

Business description — One to three sentences covering: what you do, who you serve, and where you operate. Include city names explicitly. Mention South Africa, your specific region, and major cities you serve.

Key service pages — List your most important service or product pages with brief descriptions. Keep descriptions to one line but make them specific. "AI automation services" is too vague. "AI process automation for South African financial services companies" is citable.

Contact and about page — AI models frequently cite business contact and about pages when answering "who is" and "how to contact" queries.

Recommended Additions

Blog or resources section — If you publish content regularly, list your blog with a description. This signals to AI crawlers that your site is actively updated.

Key facts block — A structured list of facts about your business: founding year, location, specializations, certifications, languages, regulatory compliance (POPIA, FSCA, HPCSA, etc.).

Team or credentials — Named team members with specific expertise signals enhance entity recognition.

Client industries — If you serve specific industries, name them. AI models use industry context when selecting sources for industry-specific queries.

What NOT to Include

• Do not include private or sensitive information
• Do not include pages you do not want AI to index
• Do not write marketing language — plain, factual descriptions work best
• Do not make claims you cannot back up on the referenced pages
• Do not list every single page — focus on the 10-20 most important

llms-full.txt: The Extended Version

Alongside the concise llms.txt, some businesses create a llms-full.txt — a more comprehensive version that includes summarized content from key pages.

The extended file structure:

# [Business Name] — Full Content Index

[Business description]

## [Page Title]
[URL] [2-4 sentence summary of the page's key content]

## [Page Title]
[URL] [Summary]

llms-full.txt takes longer to maintain but gives AI models much richer context about your content without requiring them to crawl individual pages. For businesses with rich, unique content, this extended format can significantly improve citation frequency.

Step-by-Step Implementation Guide

Step 1: Open a plain text editor — Notepad (Windows), TextEdit in plain text mode (Mac), or VS Code.

Step 2: Write your business description — One paragraph, factual, specific. Include your SA location, services, and who you serve.

Step 3: List your top pages — Start with services, then about and contact, then key blog posts or resources. No more than 20-25 entries.

Step 4: Add a key facts block — Founding year, location, specializations, certifications, languages, regulatory compliance status.

Step 5: Save as llms.txt — UTF-8 encoding, plain text format.

Step 6: Upload to your website root — The file must be accessible at yourdomain.co.za/llms.txt. This typically means uploading to the public_html folder (cPanel hosting) or the root of your site's public folder.

Step 7: Verify accessibility — Open a browser in private/incognito mode and navigate to yourdomain.co.za/llms.txt. You should see the plain text content.

Step 8: Check your GEO audit score — Run our free GEO audit to confirm your llms.txt is detected and graded correctly.

Common llms.txt Mistakes

Wrong file location — The file must be at the root domain, not in a subfolder. /llms.txt is correct; /blog/llms.txt will not be found by most crawlers.

HTML instead of plain text — Some website builders convert uploaded text files to HTML. Verify the file is served as plain text (content-type: text/plain) not text/html.

Too vague — "We provide business solutions" gives AI crawlers nothing useful. Be specific about what you actually do.

Never updated — If your services change significantly, update your llms.txt. Stale files are better than no file, but accurate files are best.

Forgetting SA context — Many South African businesses write generic llms.txt files with no geographic context. Explicitly mentioning South Africa, your provinces, and major cities you serve increases citation relevance for SA-specific queries.

Checking If AI Crawlers Are Reading Your llms.txt

You can check your server logs for user-agent strings from AI crawlers:

• OAI-SearchBot (OpenAI/ChatGPT)
• PerplexityBot (Perplexity)
• ClaudeBot (Anthropic)
• GoogleBot (Google AI Overviews, indirectly)
• Applebot-Extended (Apple Intelligence)

If you see these crawlers in your logs after adding llms.txt, they are actively reading your guidance. This typically takes 1-4 weeks after initial publication for major AI crawlers to discover your file.

Our GEO audit tool checks your llms.txt setup as part of its comprehensive scoring and flags any issues with accessibility, format, or content quality.

Maintaining Your llms.txt Over Time

llms.txt is not a "set and forget" file. Plan to review it:

• Monthly — check that all listed URLs still exist and are current
• Quarterly — update descriptions if your services or positioning have changed
• Immediately — when you launch new key service pages or significantly update existing ones

A quarterly calendar reminder for llms.txt review takes 15 minutes and keeps your AI indexing guidance accurate.

---

Related Resources:

• Free GEO Audit Tool
• GEO Optimization Service
• robots.txt for AI Crawlers: The 2026 Guide
• How to Appear in ChatGPT Answers: A Guide for SA Businesses

Further Reading:

• Search Engine Land

Frequently Asked Questions

Is llms.txt an official standard?

llms.txt was proposed by Answer.AI in 2024 and has since gained widespread adoption among AI companies and GEO practitioners. It is a de facto standard rather than an official W3C or IETF specification, but major AI systems including ChatGPT and Perplexity actively read and process it.

Does llms.txt affect my Google rankings?

No. llms.txt is read by AI language model crawlers, not Google's ranking algorithm. It has no positive or negative effect on your Google search rankings. It is purely for improving how AI answer engines understand and cite your content.

How do I add llms.txt to a WordPress website?

For WordPress, you can either upload the file directly via FTP/SFTP to your site's root directory, or use a file manager in cPanel to create the file in public_html. There are also WordPress plugins that help manage llms.txt and related AI crawler settings.

What is the difference between llms.txt and robots.txt?

robots.txt tells crawlers what they are NOT allowed to access. llms.txt tells AI crawlers what your site IS about and which content is most important. They serve opposite purposes. Both should exist on your site — robots.txt for crawler control, llms.txt for AI content guidance.

Can I have multiple llms.txt files for different sections of my site?

The standard specifies one llms.txt at the root domain. Subdomains can have their own llms.txt (e.g., blog.yourdomain.co.za/llms.txt). Subdirectories cannot have separate llms.txt files — all guidance goes in the root file.

How long should my llms.txt be?

A good llms.txt is concise and scannable — typically 30-80 lines. The llms-full.txt variant can be longer (100-300 lines) when you want to include content summaries. Do not make the file excessively long; focus on your 10-20 most important pages and key business facts.

Do I need a developer to create llms.txt?

No. Creating llms.txt requires only a plain text editor and the ability to upload a file to your website. Most website owners can do this themselves in under an hour. The challenging part is writing accurate, specific descriptions — which only you know.