← Back
AWS ML Blog

Building intent-driven UI testing with Amazon Nova Act

•12 min read•
#agents#deployment#amazon
Level:Advanced
For:AI Engineers
✦TL;DR

Epsilon built a multi‑tenant, intent‑driven UI testing service by integrating Amazon Nova Act with the Amazon Bedrock AgentCore Browser tool, layering natural‑language test steps on top of their existing Playwright regression suite. The solution runs 18‑step test workflows in managed, ephemeral browser sessions and leverages the Model Context Protocol for secure, cross‑account agent execution. This approach gives product teams a resilient, API‑driven testing capability without additional browser or identity infrastructure.

⚡ Key Takeaways

  • 18‑step test workflow demonstrated using Nova Act and Bedrock AgentCore Browser.
  • Integration uses the Model Context Protocol (MCP) for federated identity and pre‑flight authorization.
  • Managed, ephemeral browser sessions eliminate the need for each team to provision browser infrastructure, reducing maintenance overhead.
  • Teams invoke the service through the same API surface used for other agent capabilities on the platform.
  • Requires the existing Playwright regression suite as the deterministic backbone; Nova Act adds natural‑language intent steps on top.
💡 Why It Matters

Engineers can now add robust, selector‑agnostic UI tests that survive cosmetic changes while reusing their Playwright suites, accelerating release cycles without extra ops burden.

✅ Practical Steps

  1. Deploy the Amazon Nova Act service and configure it with the Amazon Bedrock AgentCore Browser tool as shown in the integration code example.
  2. Define test scenarios as plain‑English steps (e.g., “Confirm the rewards balance is visible”) and submit them via the platform’s agent API to execute the 18‑step workflow.
  3. Leverage the Model Context Protocol integration to ensure federated identity and cross‑account authorization for each test run.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

Introducing workload recovery and expanded infrastructure checks on Amazon SageMaker HyperPod

AWS ML Blog•#deployment

Introducing Unity Gateway Insights

Databricks Blog•#deployment

Introducing Gemini 4 Neon

Google DeepMind•#llm

DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency

Hugging Face Blog•#llm

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING