Node reference

Every node supports variable extraction (its outputs feed downstream steps) and a retry policy for transient errors.

Synthetic and observability

Web Test — drive a browser journey (see Synthetic → Web tests).
API Test — HTTP request(s) with assertions and extraction.
APM Observe — pull a service's recent traces; outputs span count, error flag, latency. Assert on them.

Collectors and network

DB Collector — run a read-only query against a database (via an agent that can reach it) and assert on the result.
Log Collector — pull recent log lines for a service/host and assert on presence/absence of patterns.
Network Probe — reachability/latency to a target.

Logic and actions

Assertion — boolean conditions (AND/OR groups) over any upstream value.
Branch — take different paths based on a condition.
For Each — loop over a list (e.g. run a check per region).
Approval — pause for a human to approve a risky next step.
Delay / Wait-for-HTTP — pause, or poll until an endpoint is ready.
HTTP Action — call any API to remediate (restart, scale, flip a flag).

AI

AI Analyze — an agentic session that narrates and explains the run.
AI Probe / AI Evaluate / AI Eval Suite / AI Model Compare — monitor and gate AI features (see AI Quality and Sentinel).

Full config, outputs and examples

This page is the map. For every node's form fields, outputs and worked examples, open the deep references in this category: Node reference: Synthetic and observability, Collectors and network, Logic and actions, and AI nodes.