Test Design Agent

Reads JIRA stories and Confluence design to generate risk-based test cases (functional, negative, boundary, regulatory).

RunningautonomySupervised

LangGraph pattern

Planner → Executor → Verifier

Node 1

Planner

Builds plan from JIRA/Confluence context

GPT-4.1 Mini

Node 2

Executor

Calls typed tools, drafts artifacts

GPT-5.1

Node 3

Verifier

Cross-checks output against acceptance criteria

GPT-4.1 Mini

Delivery contract

Outputs and constraints

This agent produces artifacts in JIRA, Confluence, Azure AI Search and escalates to a human reviewer at every stage gate. All runs are logged with full prompt/tool traces for audit.

State

Current

Status: Running
Autonomy: autonomySupervised
Owner: Layla Al-Mansoori
Last run: 4 min ago
Success rate: 92%
Avg tokens / run: 14,200

Environments

Available in

DEVUATPREPRODPROD