The web is changing fast. Search engines aren’t the only systems crawling websites anymore—AI models, retrieval engines, and “answer bots” now index public content to power their responses. For a site like TrailsNH, which publishes curated, geo‑tagged, real‑time hiking information, this shift matters.
To protect that work while still allowing responsible AI access, TrailsNH now uses a new standard called llms.txt. If you’ve never heard of it, you’re not alone. It’s the emerging equivalent of robots.txt—but for AI systems.
Let’s break down what it is, why TrailsNH needs one, how it works, and what outcomes we expect.
What Is an llms.txt File?
Think of llms.txt as a policy document for AI crawlers. It tells AI systems:
- what they’re allowed to index
- what they’re not allowed to use for training
- how fast they may crawl
- what attribution is required
- what content is copyrighted
- how to contact the site owner
It’s a simple text file placed at: https://trailsnh.com/llms.txt
AI companies increasingly check this file before crawling or using content. It’s not a formal standard yet, but it’s quickly becoming one.
Why TrailsNH Needs an llms.txt File
TrailsNH is built on:
- geo‑tagged data
- custom shaded relief maps
- road status information
- satellite trail overlays
- route planning tools (coming soon)
- trail descriptions (coming soon)
- thousands of pages updated daily
This is high‑value, high‑effort content.
It’s also content that AI companies would love to ingest into their models.
But TrailsNH’s mission is to serve hikers directly—not to become free training data for large AI systems.
The llms.txt file allows TrailsNH to:
1. Allow AI assistants to reference TrailsNH
Hikers can still ask AI tools about:
- trail conditions
- road closures
- weather
- trip planning
…and the AI can point back to TrailsNH.
2. Prevent AI companies from training on TrailsNH data
Training restrictions protect:
- the curated dataset
- the map layers
- the real‑time conditions summaries
- the unique value TrailsNH provides
3. Reduce server load from AI crawlers
A Crawl-delay: 30 slows polite AI crawlers to one request every 30 seconds, preventing unnecessary CPU spikes.
4. Establish clear licensing and attribution rules
If an AI system uses TrailsNH content, it must cite the source.
5. Protect the future TrailsNH chatbot
TrailsNH is building its own AI assistant. Keeping control of the data ensures that the TrailsNH bot remains the authoritative source for hikers.
How the TrailsNH llms.txt File Works
Here’s what the file tells AI crawlers:
# TrailsNH LLM Access Policy
# Last updated: 2026-03-06
# TrailsNH is a real-time hiking information system built on curated, geo-tagged data.
# This file defines how AI crawlers may access and use TrailsNH content.
# -------------------------
# Allowed AI Crawlers
# -------------------------
User-agent: *
Allow: /
# -------------------------
# Training Restrictions
# -------------------------
# AI systems may index TrailsNH content for search, retrieval, and answer generation.
# AI systems may NOT use TrailsNH content for model training, fine-tuning, embeddings training,
# dataset creation, or any derivative training processes.
Disallow-Training: /
# -------------------------
# Crawl Behavior
# -------------------------
# TrailsNH contains thousands of pages updated daily. Please crawl responsibly.
Crawl-delay: 30
# -------------------------
# Attribution Requirements
# -------------------------
# When quoting, summarizing, or referencing TrailsNH content, attribution is required:
# “Source: TrailsNH.com — Real-time hiking conditions and trip planning.”
Attribution: Required
# -------------------------
# Licensing Terms
# -------------------------
# The following content is © copyright TrailsNH.com: conditions summaries, geo-tagged data,
# hill Shaded trail map, satellite trail map, road status information, route planning, and trail descriptions.
# Trail reports and trail conditions may be © copyrighted elsewhere.
#
# AI systems may:
# - Index and retrieve TrailsNH content for user queries
# - Display short excerpts or summaries with attribution
#
# AI systems may NOT:
# - Reproduce full pages or large structured datasets
# - Redistribute TrailsNH data in bulk
# - Use TrailsNH content to train or improve AI models
# - Create derivative datasets or embeddings from TrailsNH content
License: CC-BY-NC-ND-4.0 (with additional restrictions above)
# -------------------------
# Contact
# -------------------------
Contact: https://trailsnh.com/contact/
Owner: TrailsNH.com
Purpose: Protecting curated hiking data while allowing responsible AI access.1. Indexing is allowed
AI systems may read and index TrailsNH pages for search and retrieval.
2. Training is prohibited
They may not:
- train models
- fine‑tune
- build embeddings
- create derivative datasets
- redistribute TrailsNH data
3. Crawl slowly
A 30‑second delay keeps server load predictable.
4. Attribution is required
Any AI‑generated answer must cite:
“Source: TrailsNH.com — Real-time hiking conditions and trip planning.”
5. Licensing is explicit
TrailsNH content is copyrighted, and trail reports may be copyrighted by their original authors.
6. Contact information is provided
AI companies know exactly where to reach out for permissions or questions.
Expected Outcomes
1. Lower server load
Polite AI crawlers will slow down dramatically. Combined with country blocking and human verification, TrailsNH now runs at ~25% of its resource limits instead of exceeding them.
2. Better protection of curated data
AI companies can’t legally train on TrailsNH’s unique dataset.
3. Continued visibility in AI assistants
Hikers can still ask AI tools about conditions, and the AI can reference TrailsNH without copying it.
4. Clear legal footing
If an AI system misuses TrailsNH content, the llms.txt file establishes intent and policy.
5. Stronger foundation for the TrailsNH chatbot
By controlling how external AI systems use the data, TrailsNH preserves its competitive advantage.
Final Thoughts
The llms.txt file is part of a broader shift in how the web interacts with AI. For TrailsNH, it’s a way to:
- protect years of curated work
- reduce server strain
- maintain visibility
- and prepare for the next generation of hiker tools
If you run a content‑rich website, especially one with unique or curated data, it may be time to consider your own llms.txt strategy.
