Deep scanUnclassified/Scanned 3 Sept 2026, 19:33 UTC/2 agent walks, 15 requests/Partial scan
patagonia.com
Patagonia's site returns a blank 404 to any automated browser at the homepage itself, so neither AI agents nor scanners can reach the catalog, pricing or anything else.
Two issues, in priority order
Each one is drawn from evidence recorded during the walks.
3 pieces of evidence and the recommended fix are recorded for this issue.
What a visitor actually meets
2 moments, 2 blocking
We walked the funnel in a real browser and captured every step. Nothing was submitted and no account was created.
- 01
Landing page load
Navigating to https://www.patagonia.com/ returns an HTTP 404 with body text 'Not found'. No headings, links, buttons or visible text are rendered for an automated visitor. This is the site failing to render at the entry point, not a partial or slow load.
blocked·www.patagonia.com
Patagonia landing page - 02
Attempts to find any working page
Tried root, /shop/mens, and /home/. Shop path aborted navigation entirely. /home/ returned the same blank 404 as root, no CSS, nav, text or links. Nothing here for an automated visitor to read or act on: no catalog, no pricing, no signup, just a dead 404 across paths.
blocked·/home/
Attempt to load /home/ path, also not found
We sent a buying agent
15 requests, in order
The agent got through 4 of 15 requests, but could not find out whether machines are welcome. It came up empty on 10 other checks too. That is enough missing for a machine to give up before it reaches a purchase.
- 01reached
patagonia.com
reach the site at all
- 02not found
/robots.txt
find out whether machines are welcome
- 03reached
patagonia.com
ask whether the door opens for requests carrying the AI crawlers' names
- 04not found
/llms.txt
read the site's own guide for language models
- 05not found
patagonia.com
find out what pages exist
- 06not found
/.well-known/security.txt
find a contact route for machines
- 07not found
patagonia.com
find a published agent interface
- 08not found
/.well-known/ai-plugin.json
find an agent plugin manifest
- 09not found
patagonia.com
find out whether verified agents are recognised
- 10not found
patagonia.com
read the homepage as data rather than as a page
- 11not found
patagonia.com
find a machine readable API spec
- 12not found
patagonia.com
read the developer documentation
- 13reached
patagonia.com
read the prices without running JavaScript
- 14reached
patagonia.com
walk the way a visitor would, to a signup form or a product
- 15not found
/shop/
check if any commerce/checkout path is reachable via plain HTTP
Ten dimensions
Not measured is not zero.
The agent door
9 of 100What a machine can do on your site with nobody watching.
Machine discoverability
Whether a retrieval system can find and read you at all.
No AI crawler is explicitly blocked, but a real browser gets a stripped 97-character page, there is no sitemap and no schema.org markup anywhere on the homepage. Machines cannot map what this site offers.
4 observations recorded.
Programmatic onboarding
Whether an agent can get working access without a human.
No developer documentation, no OpenAPI spec, and no signup or sign-in flow anywhere on the homepage or standard paths. The product needs no account, but there is also no API path for an agent to act on.
4 observations recorded.
Pricing legibility
Whether total cost can be computed from published numbers.
Could not verify: this scan did not reach the pages that would show it.
1 observation recorded.
Agent-aware instrumentation
Whether you publish anything built for machine callers.
No MCP manifest at any of the three standard paths, no ai-plugin.json, no OpenAPI spec at six standard paths. There is no machine-readable surface for an agent to instrument against.
4 observations recorded.
Machine-fetchable trust
Whether terms, security and status are readable as text.
No security.txt, no robots.txt (404), no sitemap declared anywhere. A machine has no published trust or crawl signal to check before acting.
2 observations recorded.
Commercial rails
Whether a purchase can complete without a sales call.
.
4 observations recorded.
The human door
94 of 100What a first time visitor meets between landing and first value.
Steps to first value
How far a new visitor travels before something useful happens.
The product is usable immediately with no account and no signup required, so there is effectively no step between arrival and use, where the site renders at all.
1 observation recorded.
Required fields
How much you ask for before you give anything back.
Nothing is asked of a visitor because there is no form on the homepage. There is no field burden to score against.
1 observation recorded.
Verification walls
How many gates stand between intent and access.
There is no account system, so there is nothing to verify. No email confirmation, no phone check, no identity gate stands between a visitor and the site.
1 observation recorded.
Error recovery
What happens when someone gets it wrong.
There is no form to submit incorrectly, so there is no error state to recover from. The score reflects the absence of a failure mode rather than a tested recovery path.
1 observation recorded.
Show the score
The badge always shows the latest scores for this domain and links back to the report. Put it in a README or a footer.
This number moves every time you deploy
SecondDoor rescans on a schedule and tells you the day a release closes a door. Watch the trend on the dashboard, or scan another site now.