AI Agent Testing: Systematically validate planning, memory, tool use, and failure recovery
Evaluate an AI agent's planning, memory, tool feedback, and stopping decisions across multi-step tasks and recovery paths.
11 min read
Blog
Page 2
PHP2Shell: How AI Helped Uncover Long-Lived PHP Vulnerabilities
A source-grounded review of the two PHP2Shell flaws, their impact and disclosure timeline, AI-assisted research, and the response expected from website and device operators.
Agent Memory Testing Skill: Keep remembered facts scoped, current, and safe to reuse
Agent Memory Testing field guide: review memory scope, retrieval, staleness, deletion, consent, and evidence across conversations.
How to Test with a Screen Reader
An introduction to common screen readers, getting started, and questions to ask while testing headings, forms, dynamic content, and links.
From Prompts to Skills: Turning Personal AI Testing Practices into Team Workflows
Follow a complete Prompt-to-Skill example and learn how sample evaluation, a quality loop, and a team pilot can capture an AI testing method.
Secrets-Exposure Review Skill: Trace Credentials Across Config, Logs, and Rotation
Trace secret exposure across configuration, logs, build output, and rotation evidence without accessing real credentials.
How to Compare Direct Chat, Prompts, and Skills on One Requirement
A reproducible comparison method with illustrative outputs for one login requirement; repeated model runs are needed to measure stability and revision time.
From Test Management to Quality Management Systems: Finding Friction in the Work System
A source-led review of Melissa Fisher and Rosie Sherry's discussion on QMS, friction mapping, product risk, quality orchestration, and AI-era collaboration, with interview views separated from engineering extensions.
API Security Testing Skill: Verify the Security Boundary from Identity to Object
Review API contracts and sanitized evidence for identity, object access, inputs, errors, and rate limits; prepare checks pending authorization.
Session-Security Testing Skill: Review Creation, Revocation, Expiry, and Device Boundaries
Review session creation, fixation, expiry, revocation, concurrent use, and cross-device security boundaries.
Can You Trust AI-Generated Test Cases? Finding Gaps and Weak Assertions
Use requirement traceability, risk coverage, and assertion review to find omissions, unsupported assumptions, unexecutable steps, and false passes in AI-generated test cases.
From Test Cases to Test Automation: Let AI Help Write and Run Tests
Use a login example to give AI project context, generate API and UI automation, and verify assertions, data isolation, and execution results.