Siftwright API docs

Version 1 · Base URL https://siftwright.com/v1 · Updated 28 September 2026

The Siftwright API gives you every Siftwright tool behind one key: YouTube transcripts, website screenshots/PDFs, Google News, company contact details, website tech stacks, App Store reviews, company job openings, domain WHOIS/DNS and post-quantum TLS scans, over REST or a hosted MCP server. Every endpoint takes a JSON body with POST and returns JSON. Get a key by subscribing to a plan, or try the tools first in the live demo.

Authentication

Send your key in the Authorization header on every request:

Authorization: Bearer sw_live_YOUR_KEY

Keys are shown once after checkout, and we store only a hash of them. Keep your key server-side, and never put it in browser or mobile app code. If a key leaks or is lost, email support@siftwright.com and we'll replace it.

Endpoints

POST /v1/youtube-transcript

Transcripts and subtitles for YouTube videos, Shorts, playlists and channels. Overview.

Counts one result per video whose transcript came back.

POST /v1/screenshot

PNG/JPEG screenshots or PDFs of web pages. Overview. Each result includes a fileUrl: a signed link you can download without your API key while the file is retained (7 days).

Counts one result per page captured.

POST /v1/google-news

Google News articles for a keyword, brand or topic. Overview.

Counts one result per article returned.

POST /v1/pqc-scan

Post-quantum TLS readiness for public domains: which key-exchange groups the server accepts (X25519MLKEM768, NIST hybrids, pure ML-KEM, the old Kyber draft, classical curves), whether it prefers post-quantum, TLS 1.2/1.3 support, the leaf certificate's algorithms and expiry, and HSTS. Same checks as the free checker.

Each result has grade (letter, label, notes), pqKeyExchange, groups, prefersPostQuantum, tls13, tls12, certificate, hsts and method: live-handshake, or inferred-cloudflare for hosts on Cloudflare's network, which our Workers can't probe directly. Results are cached for 30 minutes per host. Counts one result per host that answered; unreachable hosts are free.

POST /v1/contact-details

Public emails, phone numbers and social profiles from company websites. Overview.

Counts one result per website with "status": "ok".

POST /v1/tech-stack

CMS, ecommerce platform, frameworks, analytics, ad pixels, hosting, CDN, email and DNS provider of a website, with evidence. Overview.

Counts one result per website with "status": "ok".

POST /v1/app-store-reviews

Apple App Store reviews by app name, ID or URL. Overview.

Counts one result per 25 reviews returned, rounded up per request.

POST /v1/ats-jobs

Open jobs from Greenhouse, Lever, Ashby, SmartRecruiters and Recruitee boards in one schema. Overview.

Counts one result per 3 jobs returned, rounded up per request.

POST /v1/domain-whois

Registrar, creation and expiry dates, status, nameservers and DNSSEC from RDAP with WHOIS fallback, plus A, MX, SPF and DMARC records. Overview.

Counts one result per 3 domains with a registry answer, rounded up per request.

Response format

{
 "endpoint": "google-news",
 "count": 10,
 "billed": 10,
 "results": [ { "title": "...", "link": "...", "source": "...", "publishedAt": "...", "status": "ok" } ],
 "usage": { "plan": "pro", "used": 1210, "quota": 50000, "remaining": 48790, "period_end": "2026-10-28T12:00:00.000Z" }
}

Each item has a status. Only items with "status": "ok" count toward your plan (billed). Failed items are included so you can see why, and they're free.

Examples

curl

curl https://siftwright.com/v1/screenshot \
  -H "Authorization: Bearer $SIFTWRIGHT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com", "format": "pdf", "pdfPageFormat": "A4"}'

Python

import os, requests

r = requests.post(
    "https://siftwright.com/v1/youtube-transcript",
    headers={"Authorization": f"Bearer {os.environ['SIFTWRIGHT_API_KEY']}"},
    json={"url": "https://www.youtube.com/watch?v=jNQXAC9IVRw", "outputFormats": ["text"]},
    timeout=300,
)
r.raise_for_status()
video = r.json()["results"][0]
print(video["title"], video["wordCount"], "words")
print(video["text"][:500])

JavaScript (Node 18+)

const res = await fetch("https://siftwright.com/v1/google-news", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.SIFTWRIGHT_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ query: "artificial intelligence", maxItems: 10, country: "GB" }),
});
const { results, usage } = await res.json();
results.forEach((a) => console.log(a.source, a.title));
console.log(usage.remaining, "results left this period");

More complete, tested walk-throughs are on the blog.

Limits

• Rate limit: 60 requests per minute per key.
• Requests run synchronously and can take up to about 5 minutes for large batches, so set a generous client timeout. Small requests typically return in a few seconds.
• Usage resets at the start of each billing period. Unused results don't roll over.
• A request never uses more than your remaining results. Large batches are trimmed to fit.

Errors

Errors return a JSON body like {"error": {"code": "invalid_api_key", "message": "..."}}.

AI agents and MCP

Siftwright runs a hosted MCP server at https://siftwright.com/mcp (Streamable HTTP). Connect it with your API key and your agent gets nine tools: youtube_transcript, website_screenshot, google_news, pqc_scan, contact_details, tech_stack, app_store_reviews, ats_jobs and domain_whois. Tool calls use the same plan, quota, rate limit and billing as the REST endpoints: only successful results count. Listing the tools works without a key; calling them needs one.

Claude Code:

claude mcp add --transport http siftwright https://siftwright.com/mcp --header "Authorization: Bearer sw_live_YOUR_KEY"

Claude Desktop, Cursor, VS Code and other clients that accept a JSON config:

{
  "mcpServers": {
    "siftwright": {
      "type": "http",
      "url": "https://siftwright.com/mcp",
      "headers": { "Authorization": "Bearer sw_live_YOUR_KEY" }
    }
  }
}

No Siftwright plan? The same tools also run through Apify's hosted MCP server with an Apify account, billed pay-per-event by Apify, separately from this API:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=siftwright/youtube-transcript-extractor,siftwright/pageframe-screenshots,siftwright/google-news-scraper,siftwright/contact-details-extractor,siftwright/app-store-reviews-scraper,siftwright/tech-stack-detector,siftwright/ats-jobs-scraper,siftwright/domain-whois-dns-lookup,siftwright/email-validator-mx-disposable",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

See web data for AI agents for tool definitions and tips.

OpenAPI spec

Every /v1 endpoint, its request fields, the response envelope and the error bodies are described in an OpenAPI 3.1 file: /openapi.json. Import it into Postman, Insomnia or your code generator of choice.

Billing

Plans renew monthly through Stripe. Cancel, switch plans or update your card anytime from Manage subscription. Cancellation takes effect at the end of the paid period. Questions: support@siftwright.com.