User-agent: * Allow: / Allow: /events/ # 2026-08-27: the bare "Disallow: /e" that used to sit here was a PREFIX match, # so it silently blocked every crawler from /events/* as well. Narrowed to the # two paths it was actually meant to cover, both of them still listed below. # 2026-09-24: what those two paths ARE changed at 8104d20 and this comment did # not follow. /e/** no longer rewrites to /e.html - firebase.json sends it to # /e/index.html, the web seat shell - and /e.html stays reachable on purpose as # the QR funnel fallback. So these are two separate surfaces in the same invite # lane, not one page and its rewrite. Both carry noindex,nofollow in the page as # well: robots.txt is advisory and neither is safe on a shared link without it. # Nothing was loosened on either date; no Disallow line has changed. Disallow: /e/ Disallow: /e.html # 2026-09-27: /w/ is the browser-only door to the same web seat shell # (firebase.json rewrites it to /e/index.html). The app claims /e/ as verified app # links, so the landing's browser button points here. Same invite lane, same shield. Disallow: /w/ # 2026-09-27: /shop/ is the standalone merchant shop page, served by the # connector through a firebase.json Cloud Run rewrite. The page carries its own noindex # meta and /shop/** also gets an X-Robots-Tag header; this keeps the crawl off too. Disallow: /shop/ Disallow: /prism/ Disallow: /internal/ Disallow: /lbtech/ Disallow: /preview/ Disallow: /outreach-tracker # 2026-10-08: /claude/prompt/ is the copy-the-prompt page the connector pass email # links to. Every url there carries a personal one-use event code in its query string. # The page carries noindex,nofollow and firebase.json sends X-Robots-Tag on it too. # /claude/ itself, the assistant QR smart link, stays crawlable. Disallow: /claude/prompt/ # 2026-09-01 (DRYDOCK #52): the financial model pages were live and crawlable. # The URL stays reachable on purpose - it may already be in an investor's hands - # so this is an indexing exclusion ONLY, not an access gate. Belt and braces: # robots.txt here, a noindex,nofollow meta in both pages, and an X-Robots-Tag # header on /financials/** in firebase.json (a header a crawler cannot ignore # even if it never fetches robots.txt). # # 2026-09-02 (DRYDOCK #52, second half): crawl closure was only half the gap - # the pages were still unauthenticated at a guessable path. The model content # moved to unguessable 10-character token filenames under /financials/ (the # /care pattern); /financials/index.html and /financials/bootstrap.html now # serve a contentless stub and stay 200 so shared links never break. # DO NOT ADD THE TOKEN FILENAMES BELOW. A Disallow line publishes the exact # path it is meant to protect, and robots.txt is the first file a scraper reads. # The directory-level Disallow here already covers every file under /financials. Disallow: /financials # 2026-09-13: the TiE Los Angeles Investor Summit floor. This event is PRIVATE - it lives # in the app behind an event code and has no public event page, no listing and no sitemap # entry. The floor drawing, its manifest and the booth hero logos are hosted here only # because the app fetches them over the remote floorplan lane (conventions/tie_la_2026 # .floorplan), so they must serve - but nothing on this site links to them and they must # not be indexed. Directory-level, which covers the manifest, the image and /logos/. Disallow: /floors/ Sitemap: https://sonarconnections.com/sitemap.xml Sitemap: https://sonarconnections.com/sitemap-events.xml # 2026-09-09: each event family gets its OWN sitemap. One shared file would be # rewritten by whichever event was generated last, silently de-listing the other. Sitemap: https://sonarconnections.com/sitemap-events-sgv-megamix.xml # 2026-08-28 phase 2: the public data citation layer moved to its own host. # Its sitemap now lives at https://data.sonarconnections.com/sitemap.xml and is # declared in that host's own robots.txt. The /lb-data/ paths 301 there.