# Eagle Virtual — the free Stablecoin Freeze & Seize Tracker
#
# What is open here, and what is protected.
#
# The SITE is open to crawlers on purpose. The per-wallet check, the guides, the
# per-coin records, the leaderboard, the monthly report and its aggregate
# downloads are public on-chain fact, meant to be found and cited — indexing the
# site is the product's reach.
#
# The BULK RESTRICTED-ADDRESS LIST is not. A person checks one wallet at a time;
# the full set of flagged addresses is not ours to hand out in a single download.
# So the one file that was ever a bulk address list — /data/recent-events.json —
# is Disallowed for every crawler, while the site itself stays crawlable. Since
# 16 September 2026 that file is not published at all. It is a build file now,
# and its rows reach the world only through the table on /events.
#
# This reverses the previous "the dataset is meant to be copied" stance and the
# per-bot `Allow: /` blocks that granted the AI and answer-engine crawlers
# everything. Those crawlers are now covered by the wildcard rule below, the same
# as any other: welcome on the site, not on the bulk list. llms.txt still carries
# the framing they should quote.

# Machine-readable use terms (RSL 1.0): https://eaglevirtual.com/license.xml
# Search, AI indexing and AI input with credit; no AI training without a
# written agreement. The human version, with the credit line and the daily
# limits, is https://eaglevirtual.com/license.
#
# Every page and file on this site names that license itself, in an HTTP
# `Link: <https://eaglevirtual.com/license.xml>; rel="license";
# type="application/rsl+xml"` header, and every page also in a
# <link rel="license"> element. RSL 1.0 treats those and a robots.txt
# `License:` line as equal. The line was here until 2026-09-28; it came out
# because robots.txt checkers read it as an unknown rule.

User-agent: *
Disallow: /account
Disallow: /api/
# Internal fixtures. The gallery renders every component in every state and the
# compare page puts UI directions side by side; both are how renders get
# approved, and neither is a page. Both are noindex as well — this stops the
# fetch, the meta tag stops the listing.
Disallow: /gallery.html
Disallow: /compare.html
# The phone prototype: the page the 2026-09-20 phone pass scores its components
# on. Also an internal fixture, and also noindex — this stops the fetch, the
# meta tag stops the listing. Both spellings, because the host serves the
# extensionless form and 308s the .html one.
Disallow: /mobile-prototype
Disallow: /mobile-prototype.html
# This file stopped being published on 16 September 2026, and so did the monthly
# report files /data/eereport/items-*.json. Both are build files now and both
# answer 404. The Disallow stays, so a crawler that remembers the old URL is
# told not to ask for it. If an address-list EXPORT is ever added (a full-set
# JSON/CSV), it must be Disallowed here too, and it must itself be the
# in-window set only.
Disallow: /data/recent-events.json

# NOT disallowed, deliberately:
#
# - The per-address pages /address/<addr> stay crawlable. Each carries its own
#   robots meta and is indexable ONLY when it has a recorded event inside the
#   six-month window; an address whose events
#   are all older, or that has none, renders noindex per-page. Blocking /address/
#   here would also block the in-window pages that carry the site's SEO.
# - The aggregate downloads under /data/ — stats.public.json, the monthly report
#   JSON/CSV, the address-format reference — carry counts, never an address list,
#   and are citation assets. Left crawlable on purpose. No file under /data
#   carries an event row any more.

Sitemap: https://eaglevirtual.com/sitemap.xml