PulseAugur
EN
LIVE 23:46:51

Nearly half of audited sites use llms.txt, but many lack useful descriptions

A recent audit of 123 websites revealed that nearly half (47%) have an `llms.txt` file, a convention for providing language models with a curated map of a site's important pages. While many sites use templated or plugin-generated files, a significant portion still lack descriptive text for the links, diminishing their utility for AI models. The audit also found a notable number of dead links in these files, suggesting they are often created and then neglected. AI

IMPACT This convention could improve AI's ability to navigate and understand websites, but current implementations often lack the necessary detail to be fully effective.

RANK_REASON Analysis of a technical convention for AI search.

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Nearly half of audited sites use llms.txt, but many lack useful descriptions

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Commentary
Analysis of a technical convention for AI search.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
1 days old
Coverage has settled into its steady-state source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Kaustav Basak ·

    Almost half the sites we audited have an llms.txt. We read all 59

    <p>llms.txt is a proposed convention: a Markdown file at the root of a site that gives language models a short, curated map of its important pages. Nothing obliges anyone to support it, and as far as we know no AI search engine documents reading it. AI coding tools and agents tha…