Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for fraud analysts

Mule Account and Network Link Analysis Agent

Find the accounts linked to a confirmed fraud account and rank them by strength of evidence.

AnalysisRisk managementFinance
See how it worksGet access
Mule Account and Network Link Analysis Agent: what goes in, what the agent does and what you get

What it does

Fraud rings use many accounts that look ordinary alone, and reviewing them one at a time misses the connections. This agent starts from a confirmed fraud account and looks for shared devices, addresses, phone numbers, funding sources and transfers. It builds a link map showing how accounts connect, then tests the strength of each link: a shared device used by a family is weak, and a shared device with matching money flows is strong. It drops weak links and extends the map from strong ones. It ends with a draft case list ranked by strength of evidence. The analyst approves any hold on an account.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Confirmed fraud account 2 USES A TOOL Pull devices, addresses, phones and funding sourcesfor the account 3 USES A TOOL Find other accounts that share each attribute 4 DOES Add transfers in and out to the link map 5 DOES Rate each link as weak, medium or strong 6 CHECKS THE RESULT Is each linked account supported by at least twosignals? If not: drop weak links or look for a second signal.Back to step 4. 7 DOES Extend the map from strong links one more step 8 DOES Rank accounts and draft the case list with evidence 9 YOU APPROVE Analyst approves any hold on an account 10 RESULT Link map and ranked case list
Read the steps as a list
  1. Confirmed fraud account
  2. Pull devices, addresses, phones and funding sources for the account
  3. Find other accounts that share each attribute
  4. Add transfers in and out to the link map
  5. Rate each link as weak, medium or strong
  6. Is each linked account supported by at least two signals?If not: drop weak links or look for a second signal. Back to step 4.
  7. Extend the map from strong links one more step
  8. Rank accounts and draft the case list with evidence
  9. Analyst approves any hold on an accountThe agent waits here for your OK.
  10. Link map and ranked case list

How it decides

It keeps a link when at least two independent signals support it, and ranks accounts by the number and strength of links.

  • Treat shared devices on public networks as weak
  • Require two independent signals to keep a link
  • Extend the map at most 2 steps from the seed
  • Rank an account higher when money flows both ways with the seed

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Signals counted as strong
  • Maximum map steps (default 2)
  • Signals required (default 2)
  • Case list format

What keeps you in control

It always asks you first

  • Analyst approves any hold, freeze or closure
  • Analyst approves referral to law enforcement

Hard limits

  • Never holds or closes accounts
  • Marks every link with its evidence

It stops when

  • Done: case list delivered and reviewed
  • Stop: the map grows past a set size, narrow the signals

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensFrom one confirmed mule account, the agent found 19 accounts sharing a device, address or phone. Eleven had only a shared Wi-Fi address, which it dropped. Eight had two signals, and four of them also received transfers from the seed. It ranked those four first. The analyst approved a hold on three and reviewed the fourth.

More agents for fraud analysts

Document handling · Monitoring

Claims Document Version Agent

A complete version history of every claim document.

8 steps1 check5 jobs
Risk management · Customer support

Spam and Scam Report Triage Agent

Clear scam and spam reports quickly while keeping false removals low.

11 steps2 checks2 jobs
Analysis · Customer support

Cash Handling and Till Variance Agent

Trace till variances to likely causes quickly and spot patterns without accusing anyone.

11 steps2 checks2 jobs
Risk management · Analysis

Visa Fraud Pattern Review Agent

A short list of related-file clusters with evidence for the officer to review

10 steps2 checks2 jobs