Skip to the rows

agentsclimarketplace

Device fingerprinting

Skill yogeshg665/sleuth-fraud-investigator/skills/device-fingerprinting

Sleuth - Agent Skills that automate payment fraud investigations, plus a deterministic Python engine as the executable reference. Explainable, deterministic scoring and decisions with tokenized card references and human-review gates.

Install

npx -y skills add yogeshg665/sleuth-fraud-investigator --skill device-fingerprinting

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Flags unrecognized devices and devices shared across many accounts, a common account-takeover pattern. WHEN: "check device", "device fingerprint", "new device", "shared device", "device takeover", "unknown device risk".

SKILL.md

1.8 KB, 349 tokens by cl100k_base, as published. Nobody here has run it

Device Fingerprinting

Overview

Assesses the device used for the transaction. A device with no prior history for the account, or one shared across many distinct accounts, elevates risk.

When to Use

During the detection phase, whenever the transaction carries a device identifier and enrichment has run.

Inputs

Input	Required	Description
`transaction.device_id`	yes	Device fingerprint identifier.
`enrichment.device_known`	no	Whether the device has prior history for the account.
`enrichment.device_account_count`	no	Distinct accounts seen on the device.

Process

If device_known is false, add severity for a first-seen device.
If device_account_count is three or more, add severity that scales with the count, reflecting a device shared across many accounts.
Emit a device_risk signal with the supporting evidence.

Outputs

Zero or one RiskSignal.

Reference Implementation

src/fraud_investigator/skills/device_fingerprint.py.

Rationalizations

Excuse	Rebuttal
"Users buy new phones all the time."	A new device is low severity alone; corroboration drives the decision.
"Shared devices happen in families."	Sharing across many unrelated accounts is the pattern, not two.

Red Flags

The skill runs without enrichment having populated device features.
A missing device identifier is treated as a known device.

Verification

A shared-device or first-seen-device case produces a signal with evidence.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.