Rishtay Walay developer and agent resources

This page documents what Rishtay Walay exposes to programs — AI agents, crawlers and scripts. In short: every public page will give you clean markdown instead of HTML if you ask for it, there are index files listing what the site publishes, and the pages carry structured data. There is no public API for member data.

Ask for markdown and you will get markdown, from the same URL that serves the HTML page. This follows the acceptmarkdown.com convention, which is ordinary HTTP proactive content negotiation as defined in RFC 9110.

curl -H 'Accept: text/markdown' https://rishtaywalay.com/pricing
curl https://rishtaywalay.com/pricing.md

• Markdown responses are sent as Content-Type: text/markdown; charset=utf-8.
• Responses carry Vary: Accept, so shared caches keep the HTML and markdown variants apart.
• A request whose Accept header rules out both text/html and text/markdown gets 406 Not Acceptable.
• Locale-prefixed paths work too: /ur/pricing.md returns the Urdu markdown.

These files describe the site in a form built for machines to read.

• /llms.txt — Structured index of this site for language models (llmstxt.org format).
• /llms-full.txt — Expanded single-document context: features, plans, FAQs, branches, policies.
• /agents.md — When to use this service, what it can and cannot answer, and how to call it.
• /developers — The public JSON endpoint, content negotiation, structured data and locale coverage.
• /sitemap.xml — Every indexable URL across all 8 locales with hreflang alternates.
• /robots.txt — Crawl rules. AI crawlers are explicitly allowed on public pages.

There is no public API for profile data, and no API key programme. Profiles are published for people looking for a marriage partner; they are not offered as a dataset, and member data is not licensed for reuse. The paths under /api/ exist to serve this site's own pages and are disallowed in robots.txt.

If you are building something that needs Rishtay Walay data and the markdown representations are not enough, please write to us and describe what you are trying to do. We would rather discuss it than have you scrape the site.

Public pages carry JSON-LD structured data, so you can extract facts without parsing prose.

• Organization, WebSite and LocalBusiness on the homepage and about page.
• FAQPage for the questions answered on the homepage.
• ProfilePage on individual profile pages.
• BreadcrumbList on section pages.

Pages are published in 8 languages: en, ur, ar, hi, fr, de, es, tr. English URLs are unprefixed, for example /pricing; other locales carry a prefix, for example /ur/pricing. Every page declares hreflang alternates and the XML sitemap lists all locale variants.

If you are crawling the site, a few things will make it go better for both of us.

• Read robots.txt first. Major AI crawlers, including GPTBot, ChatGPT-User, ClaudeBot, anthropic-ai, PerplexityBot and Google-Extended, are explicitly allowed on public pages.
• Account areas and /api/ are disallowed. They hold member data and need a login.
• Prefer the markdown representation. It is a fraction of the bytes and carries no navigation, styles or scripts.
• Start from llms.txt or the XML sitemap rather than crawling blind.

A lightweight liveness endpoint, intended for uptime monitoring. Use it instead of requesting a page.

curl https://rishtaywalay.com/api/health
{"status":"ok"}

Questions about any of this, or about access for something the markdown files do not cover, are welcome.

support@rishtaywalay.com