Email verification that tells you exactly what it checked 把「到底查了什么」写清楚的邮箱验证 API
EmailCheck is an email verification API that checks an address at domain level — syntax, MX routing, disposable and role-account signals, likely typos with a correction, domain age from registration records, and a catch-all risk signal — and returns a 0–100 score with a valid / risky / invalid verdict, for one address or a batch of up to 100. It does not probe whether an individual mailbox exists: no SMTP check, and every response says so. Free tier: 100 checks a month on RapidAPI; PRO is $9/month for 10,000.
EmailCheck 是一个邮箱验证 API,在域名层检查地址——格式、MX 路由、一次性与角色账号信号、疑似拼写错误及纠正建议、来自注册记录的域名年龄、catch-all 风险信号——返回 0–100 分与 valid / risky / invalid 裁决,支持单条或一次最多 100 条批量。它不探测单个邮箱是否存在:不做 SMTP 探测,每个响应里都写明这一点。免费档:RapidAPI 上每月 100 次;PRO 为 $9/月、10,000 次。
Try it free — one address, live result免费试一下——单个地址,实时结果
Runs against the production API, no sign-up. The box below is the same check the API sells.
直接调生产 API,不用注册。这个框就是 API 本身在卖的检查。
01What it checks — and what it doesn't它查什么——以及不查什么
Checked, per address每个地址都查的
- Syntax and domain existence
- MX records — does the domain route mail at all
- Disposable-domain and role-account signals
- Likely typos, with a correction (
gmial.com → gmail.com) - Domain age in days, from RDAP registration records (null when no record exists — never estimated)
- Catch-all risk signal, labelled as a signal, not a probed conclusion
- A 0–100 score and a verdict: valid (≥80 with MX), risky (50–79, or a likely typo), invalid
- 格式与域名是否存在
- MX 记录——域名到底收不收信
- 一次性域名与角色账号信号
- 疑似拼写错误并给纠正建议(如
gmial.com → gmail.com) - 域名年龄(天),来自 RDAP 注册记录(无记录返 null,从不估算)
- Catch-all 风险信号——明说是信号,不是探测结论
- 0–100 分与裁决:valid(≥80 且有 MX)、risky(50–79,或疑似拼错)、invalid
Not checked — by design明确不查的
- Whether a specific mailbox exists. No SMTP probing: major providers reject or greylist it, so any API that "guarantees" a mailbox is overselling.
- Whether a catch-all address is a real person. On a catch-all domain, only sending mail could tell — we return the risk signal instead.
- Anything the response doesn't list. Every response carries
scope: "domain-level"and a note stating exactly what was checked.
- 某个具体邮箱是否存在。不做 SMTP 探测:主流服务商会拒绝或灰名单它,任何「保证邮箱存在」的 API 都在过度承诺。
- Catch-all 域名上的地址是不是真人。域名全收的情况下只有真发信能确认——我们给风险信号代替。
- 响应里没列的,一概没查。每个响应都带
scope: "domain-level"和一段说明,把查过的范围写死。
02How it works怎么用
- Send one address to
GET /validate?email=...(orPOST /validatewith a JSON body). - Read the verdict, the score, and the individual checks behind it — nothing is a bare true/false.
- For a list,
POST /validate/batchwith up to 100 addresses returns per-address results plus summary counts and percentages. A bad entry marks only itself; it never fails the batch.
- 把单个地址发给
GET /validate?email=...(或用 JSON body 走POST /validate)。 - 读裁决、分数和它背后的逐项检查——不是只有一个 true/false。
- 整批名单用
POST /validate/batch,一次最多 100 条,返回逐条结果加 summary 计数与百分比。坏条目只标自己,不连坐整批。
curl "https://emailcheck-api.pages.dev/validate?email=test@gmial.com"
{
"email": "test@gmial.com", "valid": false, "score": 40, "verdict": "invalid",
"checks": { "syntax": true, "domainExists": true, "mxFound": false,
"disposable": false, "roleAccount": false, "freeProvider": false },
"suggestion": "test@gmail.com", "typoLikely": true,
"domainAgeDays": 8581, "domainAgeSource": "RDAP",
"catchAllRisk": "unknown", "scope": "domain-level",
"note": "Domain-level validation only: ... Mailbox existence is not probed
(no SMTP check), so a valid result does not confirm the
mailbox exists. The score is a domain risk score."
}
Batch errors are named JSON (MISSING_EMAIL, EMPTY_BATCH, BATCH_TOO_LARGE, INVALID_JSON), so your pipeline can react to each one. Full docs and keys live on the listings below.
批量错误都是具名的 JSON(MISSING_EMAIL、EMPTY_BATCH、BATCH_TOO_LARGE、INVALID_JSON),管线可以逐个处理。完整文档与密钥在下面两个平台的产品页里。
03Pricing价格
Subscriptions run on the marketplaces — checkout, keys and billing live there, and the total at checkout is whatever the platform shows for your region, which may include local taxes. The plans, per platform:
订阅在平台上完成——结账、密钥与计费都在平台侧,结账页总价以平台按你所在地区显示的为准,可能含当地税费。各平台档位如下:
RapidAPI
| Plan档位 | Price价格 | Included含调用量 |
|---|---|---|
| BASIC | Free免费 | 100 calls/month次/月 |
| PRO | $9/month/月 | 10,000 calls/month次/月 |
| ULTRA | $29/month/月 | 50,000 calls/month次/月 |
api.market
| Plan档位 | Price价格 | Terms条件 |
|---|---|---|
| Free Trial | Free免费 | 7 days, 100 calls天,100 次 |
| Pro | $9/month/月 | See listing for quota额度见产品页 |
| Ultra | $29/month/月 | Overage $0.002 per extra call超量 $0.002/次 |
The free tiers differ on purpose: RapidAPI gives 100 calls every month; api.market gives a 7-day trial with 100 calls. A Zyla listing has also been submitted and is under review.
两处免费口径不一样,别看混:RapidAPI 是每月 100 次;api.market 是 7 天试用共 100 次。Zyla 也已提交,正在审核中。
04Questions buyers actually ask买家真正在问的问题
- What is the best email verification API?
- Why does my email verifier charge full price for "unknown" and catch-all results?
- Can a catch-all email address actually be verified?
- How accurate are email verification APIs, really?
- How much should it cost to verify 10,000 emails?
- Does an email verification API check whether the actual mailbox exists?
- How do I verify a list of emails in bulk without calling the API one by one?
- What does a "risky" verdict mean — should I still send to risky addresses?
- How do I catch email typos like gmial.com at signup?
- Why are a third of the addresses on my B2B list on catch-all domains?
05FAQ常见问题
Does EmailCheck verify that a specific mailbox exists?EmailCheck 会验证某个具体邮箱是否存在吗?
No. It checks domain mail routing (MX), registration data and risk signals at domain level. Mailbox-level probing is unreliable — major providers reject or greylist it — so we do not do it and do not claim it.
不会。它查的是域名层的邮件路由(MX)、注册数据和风险信号。邮箱级探测并不可靠——主流服务商会拒绝或灰名单它——所以我们不做,也不假装做了。
What does the score mean?分数是什么意思?
A 0–100 domain risk score. It starts at 100 and is reduced for missing MX, unknown domains, disposable domains, role accounts and likely typos; verdicts are valid (≥80 with MX present), risky (50–79, or a likely typo) and invalid.
0–100 的域名风险分。从 100 起扣:无 MX、未知域名、一次性域名、角色账号、疑似拼写错误都会扣分;裁决为 valid(≥80 且有 MX)、risky(50–79,或疑似拼错)、invalid。
How many addresses can I validate in one batch call?一次批量调用最多能验多少条?
Up to 100. A bad entry marks only itself invalid; the batch itself still succeeds, and the summary reports counts and percentages per verdict.
最多 100 条。坏条目只把自己标成 invalid,整批照常成功,summary 会给出各裁决的计数与百分比。