High Priority
Deploy /support-ai.txt Protocol
Establish a machine-readable sitemap for AI agents, detailing your customer support knowledge base hierarchy and content prioritization.
Create a `support-ai.txt` file at the root of your documentation domain, providing a brief overview of your support content's purpose (e.g., 'This is the official knowledge base for Acme Inc. customer support inquiries.').
Include markdown-style links to your most critical support documentation hubs, product FAQs, troubleshooting guides, and community forum sections.
Add a 'Known Issues' or 'Top Support Topics' section within the file to preemptively address common queries that AI assistants might be trained on or directly answer.
Configure your Customer support teams crawler protocols effortlessly.
Join 2,000+ teams scaling with AI.
High Priority
AI Assistant Selective Indexing
Fine-tune which sections of your customer support resources should be indexed and prioritized by AI crawlers like those powering chatbots or internal knowledge retrieval systems.
Implement `robots.txt` directives for specific AI user agents (e.g., `User-agent: CustomerAI Allow: /solutions/ Allow: /how-to/ Disallow: /internal-training/` to guide indexing.
Utilize meta robots tags (`<meta name='robots' content='noindex, follow'>` or specific AI agent directives) on sensitive or outdated support pages.
Monitor AI crawler activity in your web server logs to ensure key support articles are being accessed and indexed, and irrelevant content is being ignored.
Medium Priority
Structured Content for Semantic Understanding
Employ semantic HTML and structured data formats to enhance LLM scrapers' comprehension of your support article hierarchy and key information extraction.
Enclose primary troubleshooting steps or solution sections within `<article>` tags to denote their core importance.
Use `<section>` tags with descriptive `aria-label` attributes (e.g., `aria-label='Setup Guide for Feature X'`) for distinct product feature explanations within a single article.
Ensure all data presented in tables (e.g., error code mappings, compatibility matrices) uses proper `<thead>`, `<tbody>`, and `<th>` tags for accurate AI-driven data extraction.
High Priority
RAG-Ready Response Chunking
Structure your support documentation content to be optimally segmented ('chunked') for Retrieval-Augmented Generation (RAG) pipelines used in advanced AI support agents.
Isolate distinct troubleshooting procedures or solution sets within logical content blocks, ideally under 500 words each, to facilitate precise retrieval.
Minimize reliance on implicit context; explicitly restate the core problem or product within section summaries to ensure standalone relevance for RAG.
Replace ambiguous pronouns (e.g., 'it', 'this', 'they') with explicit references to product names, feature names, or error codes to improve clarity for AI context retrieval.
