Resources · Guides
Playbooks for the work you do every week
Each one ends somewhere specific rather than describing a capability.
Start here
Your first suite
From an endpoint to a first pass rate, and how to read the result.
Choosing what to test first
Twenty scenarios that matter beat two hundred that do not. How to pick them.
Writing policy rules that hold up
Practical patterns for turning support policies into ALLOW, REQUIRE and DENY rules that are testable and unambiguous.
Connecting your agent endpoint
Authentication, request and response mapping, conversation state and tool trace export, the setup details that decide test quality.
How to read a version comparison
Fixed failures, new regressions and behaviour changes, what each section means and which numbers to distrust.
Tool validation patterns
Assertions for call presence, ordering, arguments, idempotency and failure handling, with the trade-offs of each.
Rolling out agent testing across a team
Who owns scenarios, who triages failures and how to introduce gating without stalling every release in week one.
Everyday
Gating a deploy from CI
Run the suite on every pull request and fail the build when a case that must never break, breaks.
Triaging a failure
Deciding quickly whether it is the agent, the scenario, or the evidence.
Testing a webchat agent
What to do when your agent lives behind a widget rather than a documented API.
Testing refund flows end to end
A complete worked example: policies, tools, scenarios and assertions for the highest-risk support flow there is.