A

Account farm
A cluster of created or compromised accounts controlled for abuse campaigns, testing, or resale.
ASN reputation
A coarse network signal that uses autonomous system ownership and history to estimate traffic risk.

B

Behavioral cadence
The timing pattern behind clicks, scrolls, typing, navigation, and retries.
Bot score
A vendor or internal probability estimate that traffic belongs to automation or coordinated abuse.

C

Challenge fatigue
The point where legitimate users abandon because friction is repeated, opaque, or poorly placed.
Credential stuffing
Automated login attempts using username and password pairs from earlier breaches.

D

Device graph
A model that links sessions, accounts, browsers, and networks into suspected device relationships.

E

Edge enforcement
Controls applied at CDN, WAF, or reverse-proxy layers before traffic reaches the origin.
Entropy budget
The practical uniqueness available from a set of fingerprinting signals.

F

False positive
A legitimate user or request incorrectly classified as abuse.
Fingerprint drift
Natural or adversarial change in browser, device, or network attributes over time.
Friction surface
A user-facing control that raises the cost of abuse, such as a step-up, proof, or challenge.

H

Headless browser
A browser controlled without a visible UI, commonly used for automation and testing.
Honey endpoint
A route or form element intended to attract automation and mark abusive behavior.

J

JA3 fingerprint
A TLS client fingerprint built from handshake characteristics.

M

Mitigation ladder
A graduated set of responses that moves from observation to friction to blocking.

P

Proof of work
A client computation requirement used to raise the cost of high-volume requests.
Proxy rotation
Changing source IPs or networks to avoid rate limits, reputation controls, or linkage.

R

Replay attack
Reuse of captured requests, tokens, or flows without following the intended interaction path.
Residential proxy
Proxy traffic routed through consumer ISP addresses to look closer to ordinary users.
Risk holdout
A small traffic slice preserved for measurement so teams can compare mitigation outcomes.

S

Session stitching
Joining events across page views, tabs, devices, or identifiers into a coherent session.
Signal fusion
Combining weak indicators into a stronger decision model.
Step-up
An additional verification action requested only after risk crosses a threshold.

T

Token binding
Tying a token to context so it cannot be reused freely outside the expected client or flow.