LLMReady

LLMReady is a practical resource catalog for making products easier for large language models and AI agents to discover, parse, cite, and operate.

Catalog freshness: April 12, 2026.

What LLM-Ready Means

An LLM-ready product exposes reliable content, documentation, metadata, and interfaces in formats that models and agents can retrieve and interpret. The goal is not to replace human-first UX. It is to make the same product easier to understand through search, retrieval, browser automation, APIs, and agent workflows.

This catalog is for:

• Developers designing docs, APIs, structured data, and agent integrations
• Product managers deciding which LLM-readiness work belongs in scope
• Designers shaping content, IA, accessibility, and agent-friendly flows

How To Use This Catalog

Start with the category that matches the surface you own. Prefer primary docs and standards first, then use commercial tools as examples to evaluate your own requirements. Commercial listings are examples, not endorsements.

Resource Catalog

LLM-Readable Documentation

• llms.txt - A proposed convention for publishing Markdown entry points that help LLMs find concise documentation. Use it as an emerging pattern, not as a settled web standard.
• Microsoft MarkItDown - Converts files and documents into Markdown. Useful when existing source material needs a text-first form for retrieval or review.
• Docker Docs page options - A production example of documentation that offers Markdown and plain-text views. Useful as a pattern for serving both human pages and LLM-readable text.

Discoverability And Crawl Controls

• Schema.org documentation - The shared vocabulary behind many structured data implementations. Useful for making entities, products, articles, and actions explicit.
• Google structured data introduction - Practical guidance for adding structured data to pages. Useful when LLM-readiness overlaps with search and rich result hygiene.
• OpenAI publisher and developer FAQ - Guidance for ChatGPT search visibility, crawler access, robots.txt, noindex, and GPTBot training controls. Useful for separating discoverability from training opt-out decisions.

Content Transformation

• Firecrawl documentation - APIs for scraping, crawling, and extracting web content into LLM-usable formats. Useful for testing whether your public pages degrade cleanly into structured text.
• Microsoft MarkItDown - Also fits here because it converts PDFs, Office documents, images with metadata, and other inputs into Markdown. Useful for preparing legacy content for evaluation or ingestion.

AI Visibility, AEO, And GEO

• Profound - A commercial AI visibility platform. Useful as an example of how teams monitor brand and content presence in AI answer engines.
• Semrush AI Visibility Toolkit - A commercial toolkit for AI search visibility, prompt tracking, competitor research, and technical crawl checks. Useful as a comparable category reference for SEO and marketing teams.

Agent Integration Standards

• Model Context Protocol - An open protocol for connecting AI applications to tools and data sources. Useful when a product needs agent access through typed tools instead of only web pages or free-form docs.

Testing And Evaluation

• OpenAI Evals - Evaluation tooling and guidance for measuring model behavior. Useful for turning LLM-readiness claims into repeatable checks.
• promptfoo - Open-source evaluation and red-team tooling for prompts and LLM apps. Useful for CI-style checks across prompts, providers, and scenarios.
• LangSmith evaluation concepts - Evaluation and observability concepts for LLM applications. Useful for tracing and comparing agent or retrieval behavior over time.
• Braintrust documentation - Evaluation, tracing, and prompt iteration tooling. Useful for teams that need experiment tracking around LLM behavior.
• Inspect AI - Open-source evaluation framework from the UK AI Security Institute. Useful for structured evals, task definitions, scorers, and model-graded assessments.

Contribution Criteria

LLMReady is a community-driven catalog. Contributions should:

• Prefer official documentation, standards, and primary sources
• Label commercial tools as examples, not endorsements
• Include one sentence explaining why the resource matters
• Avoid uncited hype, broad claims, and generic AI commentary
• Keep the catalog useful to developers, product managers, and designers

Notes

This repo does not include a project-level llms.txt yet. Add one only after the project has a published site or documentation surface where a root llms.txt would accurately describe what agents should read.