honeyp.ai is now in Slack
Support tickets, signup reviews and spike alerts post to a channel you choose. Replies in the thread sync back to honeyp.ai.
See the Slack integrationCatch the farm, not the fan.
honeyp.ai decides how much free credit each new signup gets, in one API call, and links every account to the ones before it. Real users never notice. Account farms stop paying off.
Want a number first? Run the free audit on last month's signups. Two minutes, nothing stored.
0
Real users allowed
0
Farm accounts caught
42ms
Median decision
Every free credit is real compute. When one person opens fifty accounts, or a script opens five thousand, you pay for every one of them.
$4.4M
in compute costs from free-trial abuse that Stripe estimates it blocked at just four AI companies, in two months.
1 in 6
new signups at AI companies come from token thieves, according to Stripe's CEO, Patrick Collison.
Now in Slack, Jira and Asana.
Reviews, tickets and alerts land where your team already works, and replies come back to honeyp.ai. Every plan includes all three.
honeyp.ai is now in Jira and Confluence
Tickets and signup reviews become Jira issues that sync both ways, with an optional Confluence overview page on a schedule.
See the Jira integration
honeyp.ai is now in Asana
Tickets, reviews and spike alerts become tasks in your project. Comments and status changes sync back.
See the Asana integrationLive in an afternoon. In shadow mode first.
Nothing changes for your users until you've seen what honeyp.ai would have done on your own traffic.
Add the snippet
honeyp.js runs on your signup page and collects counts and timings, never what people type.
Ask before you grant
Your backend makes one call and gets a decision, a credit tier and the reasons in milliseconds.
See the farms
Every signup is checked against the ones before it, so accounts run by one person show up together.
# your signup handler result = honeyp.assess(account_id=user.id, email=email, ip=ip, user_agent=ua, client_token=token) grant_free_credits(user, FREE_CREDITS * result["credit_tier"]["fraction"]) # allow: 100% allow_reduced: 20% until verified review or block: 0%
A dial, not a verdict.
Most tools hand you a risk score and walk away. honeyp.ai returns an action and its reasons, and you set where the lines fall.
AllowFull free credits
MonitorFull credits, watched
ReducedSmaller until verified
ReviewHeld for a person
BlockNo free value
dana.rossi@example.org100 creditsNo risk signals. Full free tier.
f.a.r.m+14@gmail.com20 creditsSame inbox as 13 earlier signups. Verify to get the rest.
sam@mailinator.com20 creditsDisposable email domain. Verify to get the rest.
bot_8812, signed up by API with no browser0 creditsScripted signup, headless browser, a burst from one network.
- No growth tax
A flagged real user verifies and carries on instead of hitting a wall.
- Every decision explained
Plain reasons on every call. Behaviour signals alone never block, so assistive technology is safe.
- Built to fail open
If honeyp.ai is slow or down, your signups carry on as normal.
Keep your stack. honeyp.ai owns the signup.
CAPTCHAs, payment-fraud tools and bot management each guard a different moment. honeyp.ai decides how much free value every new account gets, and keeps checking in the days after.
Scroll the table sideways to compare.
| Approach | CAPTCHAs | Payment-fraud tools, like Stripe Radar | Bot management | honeyp.ai |
|---|---|---|---|---|
| When it acts | On the form | When a payment or trial reaches them | Every request | At signup and the first days after |
| What you get back | Pass or fail | A risk score, then allow or block | Block or challenge | A decision, a credit tier and the reasons |
| Works when no card is asked | Yes | Only for what reaches them | Yes | Yes |
| One person, many accounts | They pass | Scored as each payment or trial arrives | They pass | Linked and labelled in one click |
| What a flagged real user sees | A puzzle, every time | A blocked payment or trial | A challenge page | A smaller allowance until they verify |
We collect the minimum, and never sell it.
Just enough to spot abuse, and nothing we'd be uncomfortable explaining. The security page has the detail.
What we use
- The email domain and a keyed hash of the address, never the address itself
- The network prefix (/24), not the full IP address
- Counts and timings from the signup page, never what people type
- Decisions, and the labels you give us
What we never collect
- Names, addresses, phone numbers or payment details
- What people type, or screen recordings
- Cross-site tracking or advertising identifiers
- Biometric or other sensitive data
Pay for the signups we check.
Every paid plan starts with a 14-day free trial and no card. Your users' signups are never blocked because of billing.
Free trial$014 days, 5,000 signups checked, every feature
Starter$99 a month5,000 signups checked a month, then $15 per 1,000
Most popularGrowth$499 a month50,000 a month, the scanner inside your network, your brand
EnterpriseCustomCommitted volume, a security review, an SLA
See every plan and what abuse costs youCompare plans feature by feature
The questions buyers ask.
Does honeyp.ai detect AI?
No, and nobody can honestly claim to. People using AI assistants are legitimate users. We detect abuse patterns: automated signups, one person running many accounts, disposable identities and suspicious networks.
Will it hurt our conversion?
You start in shadow mode, so nothing changes until you've seen the results on your own traffic. After that, most flagged users simply verify to get their full credits instead of being blocked.
How is this different from a CAPTCHA or a payment-fraud tool?
CAPTCHAs stop simple scripts and add friction for everyone. Payment tools such as Stripe Radar act on the payments and trials that reach them. honeyp.ai decides how much free value a new account gets, and links accounts that belong to the same person.
Do we need honeyp.ai if we already use Stripe Radar?
They guard different moments, so they work side by side. Radar checks the payments and trials that reach Stripe. honeyp.ai checks every signup, whether or not it ever reaches a payment provider, and decides how much free credit it gets instead of only allowing or blocking it. Keep Radar on payments and run honeyp.ai in shadow mode beside it to see what it adds. How honeyp.ai and Stripe Radar fit together.
What if honeyp.ai is down?
The SDK fails open by default: signups continue with full credits and the response is marked as degraded. You can choose stricter behaviour if you prefer.
Do we need a cookie banner for it?
It depends on your jurisdiction and setup, so check with your legal team. honeyp.js stores nothing on the user's device, and we'll share our privacy documentation to help that conversation.
Talk to us about Enterprise.
High volume, a security review, or custom terms? Tell us a little and we'll reply within one business day.
Tell us what you need
Stop feeding the farm.
The design-partner program: a free abuse analysis of your past signups, 30 days of shadow mode, founding pricing locked for a year, and a direct line to the founder. We reply within two business days, and nothing is enforced until you've seen the numbers.
For your security review: our DPA, trust center and security overview are available now.