AI Agent Access Control: Secrets, Sign-In, Roles and Network Limits
Add AI agent access control with hashed API tokens, role-based permissions, secret management and network allowlists, and test that each limit really works.
SQA Tools Blog
Learn testing and automation, one article at a time
Free tutorials, solved practice problems and interview prep for QA and automation engineers: Python, Playwright, Selenium, SQL, Git and more.
1,677Articles
1,141Solved Python problems
7Topics
100%Free
Latest articles
1,677 articles · page 4 of 140Deploy an AI Agent to the Cloud: A Safe First Release
Learn how to deploy an AI agent to the cloud safely, with one small server, private network access, persistent storage, HTTPS and a tested rollback.
Playwright Docker Image: Package Your QA Agent
Package your QA agent with the official Playwright Docker image. Pin the version, run as a non-root user and keep secrets out of the container image.
Agent Workflow State: Save Runs, Traces and Evidence
Store agent workflow state and test evidence safely in production. Keep a record for every run, archive traces, back up data and delete old artifacts.
AI Agent as a Service: Wrap Your QA Agent in an API
Turn your QA project into an AI agent as a service, with an HTTP API, a background worker and approvals that people can give from anywhere.
AI Agent Architecture for Production: API, Queue, Workers and Storage
Design a production AI agent architecture with an API, job queue, worker, database and storage, starting small with a single SQLite-backed service.
AI Agent in Production: What Changes for a QA Agent
Learn what running an AI agent in production really means for a QA team, from shared access and failures to cost, monitoring and clear ownership.
QA AI Portfolio Project: Future Enhancements and How to Present It
Turn your agent into a strong QA AI portfolio project. Choose the right next features, write an honest README and explain your design in an interview.
AI Master Prompt for QA Projects: Use AI as Your Build Partner
Use this AI master prompt to work with any model as a build partner on your QA agent, one phase at a time, with rules that stop it…
AI Agent Acceptance Criteria: End-to-End Checklist and MVP Demo
Use clear AI agent acceptance criteria to prove your QA agent works from a fresh clone, stops safely and is ready for a short, repeatable MVP demo.
AI Agent Evaluation: Security, Reliability and a Test Pack
Run AI agent evaluation with a fixed pack of varied requirements, score correctness and safety and work through security and reliability checklists.
AI Agent CI Pipeline: GitHub Actions and Quality Metrics
Build an AI agent CI pipeline that checks every pull request, keeps model calls out of the critical path and measures whether your agent is really improving.
Preparing for a QA interview?
Take free timed mock tests with answers and explanations.