Greip
Identity & SecurityConnect AI agents to Greip, a fraud prevention and IP intelligence API. Agents look up IP reputation and geolocation, validate submitted data, and score a transaction for fraud risk before it proceeds.
What This Integration Enables
Most fraud tools hand you a verdict and keep the reasoning. Greip does the opposite, and that is the reason to reach for it: it exposes twelve narrow lookups so the rule that turns signals into an approve or decline stays in your flow where you can read it, change it, and explain it to a customer who was declined. Agents look up a single IP address for geolocation, ISP, timezone, currency, and optional proxy, VPN, and Tor flags, run the same lookup across a batch with Bulk Lookup IP Addresses, and pull reputation and abuse signals with Get IP Threats. They resolve an Autonomous System Number to its organization and registry, a card BIN to its issuing bank, brand, type, and country, an IBAN to its parsed bank and branch, a country code to reference data, and a domain to registration and reputation signals that indicate whether it is disposable or associated with abuse.
On the scoring side, Validate Email checks syntax, domain, and MX records and flags disposable, role based, and free addresses. Validate Phone Number checks format, line type, and carrier for a given ISO country code. Detect Profanity scores free text for offensive language, optionally returning the detected words, which makes it usable as a moderation input rather than an automatic takedown. Detect Payment Fraud accepts a full transaction object and returns an AI risk assessment for approve, review, or reject. Every action takes a Mode of Live or Test, and Test returns fixture data without consuming quota, which is what makes rule changes rehearsable. Two practical notes. The connector exposes no triggers: nothing in Greip starts a flow, and every check is something your flow calls at the moment it needs an answer. And Greip authenticates with the key in the request query string rather than an Authorization header, so the key can appear in access logs and URL logging middleware around the API, which is worth accounting for when you decide where this connector runs.
Without FlowRunner
With FlowRunner
Use Case Scenarios
Pre-Capture Order Review
An order arrives and the flow holds it before payment capture. The agent runs Validate Email on the buyer address, Validate Phone Number against the shipping country, Get IP Threats on the originating IP, and Lookup Card BIN on the card prefix to see whether the issuing country matches the billing country. It assembles those signals plus the amount and device details into a transaction object and calls Detect Payment Fraud. Orders that pass every check capture normally. Orders that fail on a single weak signal, such as a free email address on an otherwise clean order, do not decline automatically. They route to the risk queue with all four underlying results attached so the reviewer sees the case, not just a number.
Signup Hygiene on a Free Tier
A product with a free tier is accumulating throwaway accounts. Before a signup is provisioned, the agent runs Validate Email to flag disposable and role based addresses and Lookup Domain on the email domain to check registration age and reputation. It runs Lookup IP Address with the security module enabled to see whether the request came through a proxy or Tor exit. Accounts that trip several signals at once are held for confirmation rather than provisioned, and the agent records which signals fired so the pattern can be reviewed weekly instead of guessed at.
Moderating User Content Without Auto Removing It
A community platform receives user posts. The agent runs Detect Profanity on each submission with the detected word list enabled, and pairs it with Get IP Threats on the poster's IP. High scoring posts are held from public display and queued with both the score and the specific matched terms, because a profanity score alone cannot tell the difference between abuse and a quoted complaint. A moderator sees the post, the score, and the words that produced it, and makes the removal call. The agent has done the sorting and left the judgment where it belongs.
Human-in-Loop Highlight
The consequential moment in a fraud connector is not any single transaction. It is the rule change. Detect Payment Fraud runs the same threshold against every customer who checks out, so a tightened score cutoff or a new decline condition is not one decision, it is every decision from that moment forward, applied to people you will never hear from because they simply gave up at checkout. Greip gives you the only tool that makes this reviewable: the Mode field, which is either Live or Test, where Test returns fixture data without consuming quota. So the agent treats the promotion from Test to Live as the gate. It replays the proposed rule against a sample of recent orders in Test mode, compiles what would have changed, and posts to the risk owner: "Proposed decline threshold change on Detect Payment Fraud. Replayed against [n] recent orders in Test mode: [n] additional declines, [n] fewer reviews. Orders that would newly decline: [list with amounts and reasons]. Promote to Live, adjust, or discard?" The risk owner promotes the rule or does not. Individual transactions flow through the approved rule without a human in the path, which is the point. The rule itself never changes without one. That is human-in-the-loop placed where it costs the least and protects the most.
Agent Capabilities
12 actionsIP Intelligence
3- Lookup IP Address Retrieves geolocation and intelligence for a single IPv4 or IPv6 address: continent, country, region, city, coordinates, timezone, currency and ISP. Optional modules add proxy, VPN and Tor security flags, currency, timezone, precise location and device details.
- Bulk Lookup IP Addresses Looks up geolocation and intelligence for a batch of IP addresses in a single request, returning per IP data. Requires a plan that includes the bulk lookup feature. Used for periodic sweeps rather than per request checks.
- Get IP Threats Returns threat intelligence and reputation data for an IP address, including proxy, VPN and Tor detection and whether the address is associated with malicious or abusive activity. Used to score the risk of an incoming connection.
Data Lookup
5- Lookup ASN Retrieves details for an Autonomous System Number, including the organization name, registry, allocation country and associated IP prefixes. Used to tell a residential ISP apart from a hosting provider.
- Lookup Card BIN Resolves a payment card BIN or IIN to its issuing bank, card brand, type, level and issuing country. Used to compare the card's issuing country against the billing and shipping countries before capture.
- Lookup Country Retrieves reference data for a country by ISO 3166-1 alpha-2 code, including names, capital, population, area and phone code. Optional modules add language, flag, currency and timezone details.
- Lookup IBAN Validates an International Bank Account Number and returns the parsed bank code, branch, country and account details. Requires a plan that includes the IBAN lookup feature.
- Lookup Domain Retrieves intelligence for a domain name, including registration and reputation signals used to assess whether the domain is disposable, suspicious or associated with abuse.
Scoring
4- Validate Email Validates and scores an email address by checking syntax, domain and MX records, and flags disposable, role based, free and suspicious addresses. Returns a reachability and fraud risk assessment.
- Validate Phone Number Validates and scores a phone number for a given country by checking format, line type and carrier, and returns a fraud risk assessment. Requires the number and its ISO country code.
- Detect Profanity Analyzes free text for profanity and offensive language and returns a score, optionally including the list of detected words. Used as a moderation input that a person acts on, not an automatic takedown.
- Detect Payment Fraud Submits a transaction object including action, amount, currency, customer and device details to Greip's AI payment fraud module and returns a risk assessment used to approve, review or reject. Every call carries a Mode of Live or Test, and rule changes are rehearsed in Test before promotion.
Frequently Asked Questions
What can FlowRunner do with Greip?
FlowRunner agents can run Lookup IP Address, Bulk Lookup IP Addresses, and Get IP Threats in Greip, plus 9 more actions.
Does connecting Greip to FlowRunner require OAuth?
No. Greip connects to FlowRunner with an API key, no OAuth flow required.
Can Greip trigger a FlowRunner workflow automatically?
Greip doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Start building with Greip
$100 in credits. No card required. Connect in minutes.