promptly3518 walls its fork off from search engines

A belt-and-braces move to keep a private legal-AI deployment out of Google's index.

securityinfrastructure

promptly3518 added two layers to stop search engines from finding and indexing the fork's web pages. The first is the standard polite request that well-behaved crawlers honour. The second is a page-level instruction baked into the app itself, which catches crawlers that ignore the polite request and reach the page anyway.

The owner frames this as defence in depth on top of access controls already enforced by Vercel, the platform hosting the deployment. The thinking: if a link ever leaks past those controls, these directives are a second line keeping the content from surfacing in public search results. The page-level instruction is the more reliable of the two and the easy piece for others to copy.

So what Anyone running a confidential legal tool on a public host should treat "don't get indexed" as a deliberate setting, not an assumption.

View this fork on GitHub →

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

1 commit from promptly3518/mike, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
39362a08 Add noindex meta + robots.txt to frontend promptly3518 2026-05-03 ↗ GitHub
commit body
Defense in depth alongside Vercel's deployment protection. Even if a URL
is somehow surfaced, search engines see noindex/nofollow plus a
Disallow-all robots.txt.

robots.txt is force-added because the global gitignore excludes public/.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-78.md from inside the repo you want the changes in.

⬇ Download capture-thread-78.md