Source Verification API & Scraper
The Agent Source API verifies and resolves sources for AI agents as clean JSON.
🤖 Using an AI assistant? Copy this link into ChatGPT / Claude / Cursor — it reads every endpoint and parameter instantly and tells you if this API fits your use case.
The primary source_url endpoint resolves a URL (final URL, domain, accessibility, status, redirect chain), and you can find an official_source, verify a claim, look up and compare pricing, and list providers. It is built for AI agents that must cite, verify and trust the sources they use. One ReefAPI key, one shared credit pool, the standard envelope.
The curated provider whitelist, measured provider by provider
official_source and the pricing actions answer only for a fixed whitelist, which the providers action returns in full. All eight resolve official URLs, but structured pricing rows are a separate question, and the parse surface behind each provider determines whether you get rows at all. The pricing column below is what a live call returned on 2026-08-26, not what the capability flag claims.
| Provider key | Official URLs | pricing_lookup measured 2026-08-26 | Parse surface |
|---|---|---|---|
| anthropic | Yes, 6 need types | 13 model rows | html_proximity |
| gemini | Yes, 6 need types | Rows returned, with the same model name repeating once per price tier | html_table |
| openrouter | Yes, 6 need types | Rows returned, with output rates that are not a fixed multiple of input, e.g. 0.16 in and 0.47 out | json_api |
| openai | Yes, 6 need types | pricing_available false, count 0 | html_proximity |
| xai | Yes, 6 need types | pricing_available false, count 0 | html_proximity |
| aws, gcp, azure | Yes, 6 need types | pricing_available false by design, with normalization_notes explaining that model prices live in sprawling compute-SKU pages | unstructured |
The six need_type values are pricing, docs, changelog, status, legal and model_card, plus 'all' to get every URL at once. Aliases are generous: 'claude', 'opus' and 'gpt-5' all resolve to their provider key. An entity outside the whitelist is not an error, it returns ok:true with matched:false, official_url null and a suggestion listing the supported keys.
Real request and response JSON
Captured from the indexed primary action, source_url, on .
{
"method": "POST",
"url": "https://api.reefapi.com/agent-source/v1/source_url",
"headers": {
"x-api-key": "$REEF_KEY",
"content-type": "application/json"
},
"body": {
"url": "https://openai.com/api/pricing/",
"official_for": "openai"
}
}{
"ok": true,
"meta": {
"api": "agent-source",
"endpoint": "source_url",
"mode": "live",
"latency_ms": 4246.7,
"record_count": 1,
"bytes": 0,
"cache_hit": false
},
"data": {
"url": "https://openai.com/api/pricing/",
"final_url": "https://openai.com/api/pricing/",
"final_domain": "openai.com",
"accessible": true,
"status_code": 403,
"blocked": false,
"redirect": {
"chain": [
{
"url": "https://openai.com/api/pricing/",
"status": 403
}
],
"hops": 0,
"redirected": false,
"loop": false
},
"metadata": {
"title": "Business Pricing | OpenAI",
"site_name": "OpenAI",
"description": "Discover OpenAI pricing for ChatGPT Business and APIs, with GPT token costs, enterprise features, developer plans, and AI tools.",
"lang": "en-US",
"canonical": "https://openai.com/business/pricing/",
"author": null,
"published_at": null,
"updated_at": null,
"og_type": "website",
"json_ld_types": [],
"favicon": "/favicon.ico?favicon.0w1tl_z9koc07.ico",
"meta_robots": "index, follow"
},
"render_mode": "bare-dc",
"official": {
"provider": "openai",
"is_official": true,
"reason": "final domain is in the provider's official-domain set"
},
"freshness": {
"updated_at": null,
"published_at": null,
"archive": {
"available": true,
"closest": "[redacted-phone]T14:56:02Z",
"snapshot_url": "http://web.archive.org/web/[redacted-phone]/https://openai.com/api/pricing/",
"age_days": 32
},
"staleness": "archived_recently"
},
"subcalls": [
{
"api": "url-resolver",
"action": "resolve",
"status": "ok",
"latency_ms": 290.4
},
{
"api": "web-extract",
"action": "scrape",
"status": "fail:TARGET_BLOCKED",
"latency_ms": 1110.6
},
{
"api": "ext:page",
"action": "fetch",
"status": "ok",
"latency_ms": 1305.3
}
]
}
}What the Source Verification API does
| Action | Description | Concrete use case | Key params |
|---|---|---|---|
| source_url | Inspect a public URL for source-reliability: reachability, canonical + full redirect chain, title/publisher/published+updated dates, language, official-likelihood, and a stale/archived (wayback) signal. | Ops teams call source_url to get inspect a public URL for source-reliability. | url, check_archive, official_for |
| official_source | Resolve the OFFICIAL URL a curated provider publishes for a given need (pricing/docs/changelog/status/legal/model_card) + why it is official (domain match + domain age via whois). Curated whitelist only. | Developer tools call official_source to resolve the OFFICIAL URL a curated provider publishes for a given need (pricing/docs/changelo…. | entity, need_type, verify |
| verify_claim | Check whether one or more public source URLs support a factual claim — returns supported, contradicted, not_found, or unclear, along with the matching evidence snippet and a confidence level. Uses exact quote and number matching against the live page content. A source-evidence checker, not a general truth oracle. | Validation workflows call verify_claim to check whether one or more public source URLs support a factual claim. | claim, url, urls |
| pricing_lookup | Current developer/LLM pricing for a curated provider, parsed live from the provider's own pricing page/API. Each row: model, input/output per-1M-token USD, source_url, last_seen_at, unit, currency, normalization_notes. | Data-quality teams call pricing_lookup to get current developer/LLM pricing for a curated provider, parsed live from the provider's own pri…. | provider, model, limit |
| pricing_compare | Compare current pricing across several curated providers (optionally filtered to a model substring) → normalized per-1M-token USD, cheapest-first, each traceable to its source_url + last_seen_at. | Ops teams call pricing_compare to compare current pricing across several curated providers (optionally filtered to a model subs…. | providers, model |
| providers | List the curated provider whitelist + each provider's capabilities (which need_types and structured pricing are supported) and parse-surface metadata. The honest denominator for official_source / pricing. | Developer tools call providers to list the curated provider whitelist + each provider's capabilities (which need_types and stru…. | provider |
Call source_url from your stack
curl -X POST https://api.reefapi.com/agent-source/v1/source_url \
-H "x-api-key: $REEF_KEY" \
-H "content-type: application/json" \
-d '{"url":"https://openai.com/api/pricing/","official_for":"openai"}'import requests
r = requests.post(
"https://api.reefapi.com/agent-source/v1/source_url",
headers={"x-api-key": REEF_KEY},
json={
"url": "https://openai.com/api/pricing/",
"official_for": "openai"
},
)
print(r.json()["data"])const res = await fetch("https://api.reefapi.com/agent-source/v1/source_url", {
method: "POST",
headers: {
"x-api-key": process.env.REEF_KEY,
"content-type": "application/json",
},
body: JSON.stringify({
"url": "https://openai.com/api/pricing/",
"official_for": "openai"
}),
});
const { ok, data, meta, error } = await res.json();Ask your MCP-connected assistant: call reefapi.agent-source.source_url with {"url":"https://openai.com/api/pricing/","official_for":"openai"}.Who uses this API and why
- AI agents call verify_claim to check a statement against an official source.
- Agents use source_url to resolve and validate a canonical link before citing it.
- Pricing agents use pricing_lookup and pricing_compare to ground cost claims.
Questions developers ask before integrating
Does a 'supported' verdict from verify_claim mean the claim is true?
No. It means the phrase and the number both appear on that page, close together. Two measured calls against the same Anthropic pricing page prove the limit: 'Claude Opus 4.1 input is $15 per million tokens' returned supported at confidence 0.9, and 'Claude Opus 4.1 output is $30 per million tokens' also returned supported at 0.9, because $30 sits in the same price row even though it is not the output rate. The payload says it plainly in its own note: this is a source-evidence check, not a truth oracle. Always read the evidence snippets, which show exactly which text was matched.
How reliable is output_per_1m_usd?
It depends on the parse surface, and the response tells you which one was used. On providers parsed with html_proximity, normalization_notes states that input is the first price found and output is inferred from a 1.5x-to-8x band, which can pick the wrong column on a page that lists cache and batch tiers side by side. Every row also carries prices_found_usd with all the nearby rates, so on a measured Opus 4.1 row you can see the full set 15, 18.75, 30, 1.5 and 75. Bill from prices_found_usd and source_url, not from the single output number.
Why does the same model appear several times in pricing rows?
Because a provider's pricing table lists one row per tier, and the parser keeps them all rather than picking a winner. A measured comparison filtered to 'flash' returned gemini-3.1-flash repeatedly, at four different input rates: 0.125, 0.25, 0.45 and 1.5 per million tokens, which are the context and modality tiers listed on that page. cheapest_input and cheapest_output pick the lowest row, so they will report the cheapest tier of a model and not its standard rate. De-duplicate on model plus price before you show anything to a user.
What happens when I ask about a provider that is not on the list?
You get an honest empty answer rather than a guess. A measured official_source call for 'some-random-startup-xyz' returned ok:true with provider null, matched false, official_url null, an empty all_sources object and a suggestion listing the eight supported keys. The disclaimer in the payload states that finding the official source for any entity on the internet is deliberately not claimed, and that the whitelist only grows with maintainer sign-off.
official_source says reachable:false but the page opens in my browser. Which is right?
The page is fine. verify:true on official_source runs a lightweight reachability probe that can report false on hosts that answer redirects unusually. A measured source_url call on the very same URL, https://openai.com/api/pricing/, returned accessible true with status_code 200 after two hops. When reachability matters, use source_url, which gives you the full redirect chain with a status code per hop.
Why did source_url return a different URL than the one I sent?
It follows the redirect chain to the end, including any locale rewrite the site performs on your request. The measured chain from https://openai.com/api/pricing/ was 308 to https://openai.com/business/pricing/#api, then 302 to https://openai.com/tr-TR/business/pricing/, which returned 200. Read final_url and final_domain rather than the URL you sent, and note that official.is_official is decided on the final domain, which is why that redirected page still scored true for openai.
The freshness block says staleness 'unknown'. What should I read instead?
It means the page published neither a date nor a usable archive snapshot. In a measured call, metadata.published_at, updated_at, title and canonical were all null, archive.available was false, and staleness fell back to 'unknown'. Modern marketing pages frequently ship no dateable metadata at all. When you need a timestamp on a price, use last_seen_at from pricing_lookup, which records when the page was actually read, rather than trying to infer the page's own age.
How do I get an official URL for AWS, GCP or Azure pricing?
Use official_source, which resolves them normally, and skip pricing_lookup for those three. A measured pricing_lookup on aws returned ok:true with pricing_available false, an empty models array, and normalization_notes explaining that model pricing there lives inside compute-SKU pages with no stable structured slice, so it is out of scope. The official pricing URL still comes back, along with why_official showing the domain match, its age in days and the registrar.
What is the Source Verification API?
Source Verification API is a ReefAPI endpoint group for source verification It returns live JSON through POST requests under /agent-source/v1.
Is the Source Verification API free to try?
Yes. ReefAPI starts with 1,000 free credits, no card required. Source Verification calls use the same shared credit balance as every other ReefAPI engine.
Do I need a Source Verification login or account?
No login to Source Verification is needed for the API response. You call ReefAPI with your x-api-key header, and the playground can run live examples before you create a production key.
How fresh is the Source Verification data?
The page example is captured from a live source_url call, and production requests fetch live data through ReefAPI rather than a static sample.
How many credits does the Source Verification API use?
Source Verification actions currently cost 1-3 credits per successful call. Failed or blocked calls are free, and all APIs draw from one credit pool.
Can I call Source Verification from an AI assistant or MCP client?
Yes. Connect ReefAPI once through MCP and your assistant can call agent-source actions with the same key, credit pool and JSON envelope used by normal REST requests.