timwukp

Harness Agentic AI Agent Best Practices And Use Case — Testing skill for Claude Code

Testing community

Production-ready AI agent for UI/web testing on Amazon Bedrock AgentCore Harness — Memory, Skills, observability + Bug-Fix Agent + best practices.

How to install Harness Agentic AI Agent Best Practices And Use Case

This entry records only its repository, not the path inside it, so there is no exact command to give. Open timwukp/Harness-agentic-AI-agent-best-practices-and-use-case and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Harness Agentic AI Agent Best Practices And Use Case does

Production-ready AI agent for UI/web testing on Amazon Bedrock AgentCore Harness — Memory, Skills, observability + Bug-Fix Agent + best practices.

Alternatives in Testing

Webapp Testing — Test local web applications using Playwright for UI verification and debugging 94.1k ★
Fix Issue — by metabase - Addresses GitHub issues by taking issue number as parameter, analyzing context, implementing sol 46.5k ★
AI Contribution Skills (Engram) — branch-pr: clean branch and PR workflow 2.5k ★

README

AWS Bedrock AgentCore Harness — UI Test Agent

An AI agent that tests your web app like a human QA tester — then fixes the bugs it finds.
Built on Amazon Bedrock AgentCore. Runs in CI on every pull request.

🎬 Watch it work — 9 minutes, narrated

https://github.com/user-attachments/assets/e25d7708-1a75-49f5-ad90-7f7797c11306

A full walkthrough of the autonomous QA loop running against a **live production app** — no slides, no mock data. Every number on screen is read from the real AWS account. ([Also on the wiki](https://github.com/timwukp/Harness-agentic-AI-agent-best-practices-and-use-case/wiki).)

	Chapter	What you see
`0:00`	Use case	The app under test, and the QA debt this replaces
`0:56`	Architecture	The five verified stages, then inside both harnesses
`2:24`	Pipeline	`ui-qa-agent.yml` stage by stage — deploy then test, and the three brakes that stop a runaway loop
`4:16`	Observability	Live invocation counts, latency, token usage, error rate
`5:06`	Evaluation	Three online evaluators scoring the agent's own output
`5:52`	Optimization & cost	Insights → recommendations → prompt drafts; billed vs. estimated, kept separate
`7:46`	Results & value	A real finding, 66 evidence screenshots, and what the loop is worth