REST API

Base URL https://essentials.sh · no authentication · CORS enabled · full spec at /openapi.json.

Errors are structured JSON with a stable code and a hint:

{ "error": { "code": "invalid_url", "message": "…", "hint": "…" } }

Endpoints

Generate a QR code

curl "https://essentials.sh/api/qr.svg?url=https%3A%2F%2Fexample.com&ec=M" -o qr.svg

GET /api/qr.svg — returns image/svg+xml. Query params: url (required), ec (error correction: L ~7%, M ~15% default, Q ~25%, H ~30%).

Build meta tags

curl -X POST https://essentials.sh/api/meta \
  -H 'content-type: application/json' \
  -d '{"title": "Acme — ship faster", "description": "Acme helps teams ship.", "url": "https://acme.com"}'

POST /api/meta — returns { meta_tags }: title, description, canonical, Open Graph, and Twitter card tags, HTML-escaped. Optional fields: siteName, image, twitterHandle, themeColor.

Build robots.txt

curl -X POST https://essentials.sh/api/robots \
  -H 'content-type: application/json' \
  -d '{"siteUrl": "https://acme.com", "aiTrain": false, "blockAiTrainers": true}'

POST /api/robots — returns { robots_txt } with AI Content Signals (search/aiInput/aiTrain booleans), optional disallow paths, and optional hard blocks for known AI-training crawlers.

Build a sitemap

curl -X POST https://essentials.sh/api/sitemap \
  -H 'content-type: application/json' \
  -d '{"siteUrl": "https://acme.com", "paths": ["/", "/about", "/pricing"]}'

POST /api/sitemap — returns { sitemap_xml }. Optional lastmod (YYYY-MM-DD).

Build an llms.txt

curl -X POST https://essentials.sh/api/llms-txt \
  -H 'content-type: application/json' \
  -d '{"siteName": "Acme", "summary": "Ship faster.", "sections": [{"title": "Docs", "links": [{"title": "Guide", "url": "https://acme.com/docs"}]}]}'

POST /api/llms-txt — returns { llms_txt } per llmstxt.org.

Scan a site

curl "https://essentials.sh/api/scan?url=example.com"

GET /api/scan — fetches a public site and reports 12 checks (title, description, canonical, Open Graph, Twitter card, favicon, robots.txt, Content Signals, sitemap, llms.txt, JSON-LD, reachability), each failing check with a fixUrl to the tool that fixes it. Public domains only; results cached ~10 minutes per origin.