An edge layer that gives Webflow SEO and AI-search control it doesn't have natively
Webflow publishes pages well, but not edge behavior
Webflow can publish strong pages, but technical SEO and generative-engine optimization often need extra control over redirects, discovery files, headers, cache behavior, and structured route lists that live below the page level.
Those details matter more as both search engines and AI answer engines increasingly decide what to crawl, cite, and cache based on signals Webflow's page settings don't expose.
Add an edge layer without disturbing the Webflow workflow
The goal was a Cloudflare Worker that proxies the Webflow origin transparently: redirecting alternate hosts to the canonical domain, stripping tracking parameters, serving robots.txt, sitemap.xml, llms.txt, and ai.txt, and applying route-specific cache and SEO rules.
None of this could change how content gets edited or published in Webflow day to day.
A proxy that adds edge-level SEO and AI-search control
The proxy gives SoFlow's own site a stronger technical foundation for search visibility, AI-search readiness, and ongoing maintenance — all without giving up Webflow's editing experience.
Our other projects
The brief was to keep Webflow as SoFlow's marketing site and CMS while adding real control over how search engines, AI systems, and monitoring scripts understand the domain — control that normally requires leaving the platform.
Webflow can publish strong pages, but technical SEO and generative-engine optimization often need extra control over redirects, discovery files, headers, cache behavior, and structured route lists.
Those details matter when a site wants to be legible to both traditional search crawlers and newer AI answer engines at the same time.
The proxy had to preserve the existing Webflow editing workflow, avoid breaking any live page, support canonical host and URL cleanup, and keep generated or private routes — like audit reports — out of public discovery surfaces.
SoFlow implemented a Cloudflare Worker that proxies its Webflow origin: it redirects alternate hosts to the canonical domain, strips tracking parameters from URLs, serves robots.txt, sitemap.xml, llms.txt, and ai.txt directly from the edge, and applies route-specific cache and SEO handling.
Webflow keeps rendering and serving every page exactly as authored — the Worker only intercepts the surrounding technical behavior.
Visitors and crawlers request the public SoFlow domain. Cloudflare routes that request through the Worker first, which resolves canonical redirects, serves discovery files like robots.txt and llms.txt, and applies cache and header rules, before passing ordinary page requests through to the Webflow origin unchanged.
- Canonical host redirects and tracking-parameter cleanup
- Worker-served robots.txt and sitemap.xml
- llms.txt and ai.txt for AI-search discovery and citation
- Route-specific cache behavior and purge control
- Live verification checks for key public routes
The proxy gives SoFlow a stronger technical foundation for its own Webflow SEO, AI-search readiness, and ongoing site maintenance — and a reference pattern to offer clients who need the same layer.
- Serves robots.txt, sitemap.xml, llms.txt, and ai.txt directly from the Worker
- Redirects www and Webflow-origin requests to a single canonical host
- Strips common tracking parameters before URLs are treated as canonical
- Keeps private, generated audit-report routes out of public discovery
Isn't this over-engineering basic Webflow SEO settings?
For most sites, no — Webflow's built-in SEO fields, redirects, and sitemap cover the basics well. This proxy exists for the layer above that: canonical host cleanup across domain variants, a machine-readable llms.txt for AI answer engines, cache and header control, and keeping private generated routes out of public discovery. Those are edge-network concerns Webflow's page settings were never built to own.
What is llms.txt used for?
It gives AI systems a concise, structured map of the company, topics, services, and preferred citation pages. It doesn't replace strong page content — it clarifies site structure for answer engines the way a sitemap does for search crawlers.
Can this be added without rebuilding a Webflow site?
Yes. The proxy sits at the domain edge in front of the existing Webflow origin, so the editing workflow, page builder, and publishing flow stay exactly the same. Only routing, headers, and a handful of generated files change.
This build proves SoFlow can keep Webflow flexible for marketers and editors while layering in the infrastructure-level controls that SEO, AI search, and technical monitoring actually require.
Isn't this over-engineering basic Webflow SEO settings?
For most sites, no — Webflow's built-in SEO fields, redirects, and sitemap cover the basics well. This proxy exists for the layer above that: canonical host cleanup across domain variants, a machine-readable llms.txt for AI answer engines, cache and header control, and keeping private generated routes out of public discovery. Those are edge-network concerns Webflow's page settings were never built to own.
What is llms.txt used for?
Heading 1
Heading 2
Heading 3
Heading 4
Heading 5
Heading 6
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
Block quote
Ordered list
- Item 1
- Item 2
- Item 3
Unordered list
- Item A
- Item B
- Item C
Bold text
Emphasis
Superscript
Subscript
Heading 1
Heading 2
Heading 3
Heading 4
Heading 5
Heading 6
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
Block quote
Ordered list
- Item 1
- Item 2
- Item 3
Unordered list
- Item A
- Item B
- Item C
Bold text
Emphasis
Superscript
Subscript
Before the proxy, SoFlow's technical discovery behavior depended on Webflow's default surface plus separate, manual checks whenever canonical or crawling issues came up.
After the proxy, SoFlow keeps Webflow as the visual CMS while owning discovery files, canonical behavior, cache rules, and exclusions for generated routes at the edge.
Webflow's page-level SEO settings can't own edge concerns like robots.txt and sitemap.xml generation, llms.txt for AI systems, canonical URL normalization across domain variants, cache-purge behavior, or route-level uptime verification.
A proxy lets SoFlow add all of that at the network edge without touching the Webflow build itself.
