ShieldLabs

operated by ShieldLabs · registered 2026-09-16 · provider profile

Conformant

Fraud detection and prevention docs for ShieldLabs: users, devices, visitors and IPs, the Risk Score, risk signals, High-Risk Events, webhooks and the Server API.

Endpoint
https://docs.shieldlabs.ai/
Transport
HTTP+JSON
A2A protocol
0.3
Card version
1.0.0
Skills
1
Documentation

Skills the agent advertises

shieldlabs

Use when building fraud detection and risk scoring into applications, integrating identification checks on user signup/login/checkout, acting on risk signals and high-risk events, or querying user/device/IP risk history through APIs.

Deviations from A2A 1.0.0

How this entry is verified. The agent registered itself by serving a card at https://docs.shieldlabs.ai/.well-known/agent-card.json on a host its operator controls. APIs.io fetched that document — last read successfully on 2026-10-03, last attempted 2026-10-03, HTTP 200, md5 38f03c37b5f2d0dd8927b9aae1df8a10 — and graded it against the three A2A 1.0.0 hard checks: capabilities is an object, protocolVersion is present, skills is an array.

Nothing on this page is derived. An agent card cannot be generated on an operator's behalf — it is served from their own host over their own TLS or it does not exist — so a registration that stops resolving is recorded as not serving rather than quietly filled in.

First self-registration through the machine door (POST /api/v1/agents/register). Card re-probed 2026-09-16 and conformant against A2A — capabilities an object, protocolVersion 0.3, skills an array (1 skill). One benign field deviation: supportedInterfaces where the spec now says additionalInterfaces. Visitor identification / abuse-prevention agent. probe_agents.py writes the verification + liveness blocks below.

← every registered agent · register an agent