Key Concepts Glossary

This page defines the core terminology used throughout BugBrain. Understanding these concepts will help you get the most from the platform.

Test Case

A test case is a single automated scenario that verifies a specific piece of functionality. It contains:

Title — What the test does (e.g., “User can log in with valid credentials”)
Steps — Ordered instructions: navigate, click, type, assert, wait
Expected Results — What should happen after all steps execute
Tags — Categorize tests (e.g., auth, checkout, payment)
Priority — Critical, High, Medium, Low
Status — Draft, Active, or Archived

Example: “Complete checkout flow” has 8 steps: open cart → click checkout → enter email → select address → enter payment → click confirm → verify order confirmation page

Execution

An execution is a single run of a test case. It includes:

Start time and duration
Status — Passed, Failed, or Running
Steps results — Did each step succeed?
Screenshots — Captured at each step
AI failure analysis — Why it failed (if failed)
Duration and logs — How long it took and detailed execution records

Each execution is recorded in your test history so you can track trends (pass rate over time, flaky tests, etc.).

Persona

A persona is a saved login profile with credentials. Instead of typing passwords during tests, the AI uses personas to authenticate automatically. Types include:

Basic Auth — Username and password
Email OTP — Email address + OTP code via Mailosaur
Magic Link — Passwordless link sent to email
TOTP / MFA — Password + TOTP 2FA seed (Google Authenticator)

Credentials are encrypted at rest (AES-256-GCM) and never exposed in logs or screenshots.

Example: Create a “QA Admin” persona with credentials qa@example.com / SecurePass123. Use it in tests that require admin access.

Project

A project groups related tests and configurations:

Name — e.g., “Acme E-commerce”
Base URL — The application under test (e.g., https://app.example.com)
Personas — Auth profiles specific to this project
Device presets — Saved mobile/tablet/desktop configurations
Team members — Who can edit tests in this project

Projects enable multi-team collaboration and scoped access control.

Test Plan

A test plan is a collection of test cases that run together:

Type — Smoke (fast, critical paths), Regression (full suite), Sanity, or Custom
Test case list — Which tests to include
Schedule — Run on a cadence (daily, weekly, etc.) or manually
Provider — Which execution service to use

Run all tests in a plan with a single click or via CI/CD webhook.

Example: “Pre-production smoke test” includes 5 critical path tests. Runs automatically every morning at 9 AM.

Discovery Session

A discovery session crawls your application to automatically uncover:

Pages — All discoverable URLs
User flows — Authentication, CRUD operations, search, navigation
UI patterns — Forms, tables, modals, navigation menus
API endpoints — REST and GraphQL calls made by the application

No test code written. The crawler uses Playwright to explore at 50% faster speed than manual testing.

Exploratory Session

An exploratory session uses AI to autonomously test your application for bugs, without predefined test cases. The AI:

Scouts — Discovers pages and navigation
Explores — Interacts with UI elements, forms, buttons
Reports — Lists bugs found with severity scores

Especially useful for finding edge cases and UX issues humans would miss.

Compliance Audit

A compliance audit automatically checks your application against frameworks:

WCAG 2.1 — Web Content Accessibility Guidelines (color contrast, ARIA labels, keyboard navigation)
GDPR — Cookie consent, privacy policies, data subject rights forms
SOC2 — Authentication controls, session management, security headers

Violations are scored by severity (Critical, Serious, Moderate, Minor) and include remediation guidance.

Knowledge Map

A knowledge map accumulates learning about your application across multiple exploratory sessions. It tracks:

Navigation graph — Pages discovered and links between them
Coverage score — 50% (low confidence) → 100% (high confidence) as more sessions run
Risk heatmap — Which pages/features have the most potential issues
Page coverage — Detailed metrics per page

Use knowledge maps to improve AI test generation and prioritize testing efforts.

AI Agent

The AI agent is the intelligent system that:

Understands your test case description (natural language)
Navigates your application autonomously
Makes decisions about what to click, type, and assert
Learns from failures and adapts
Generates failure analysis explaining why tests failed

The agent runs up to 50 iterations per test (configurable) and can handle complex flows including authentication and dynamic content.

Table of Concepts

Concept	Purpose	Contains
Test Case	Single test scenario	Steps, title, tags, priority
Execution	One test run	Status, duration, screenshots, analysis
Test Plan	Bundle of tests	List of test cases, schedule
Project	Organizational unit	Tests, personas, team members
Persona	Login profile	Credentials (encrypted), auth type
Discovery	Web crawling	Pages, flows, UI patterns, APIs
Exploratory	Autonomous bug finding	Bugs found, severity, confidence
Compliance	Standards checking	Violations, remediation, score
Knowledge Map	Learning accumulation	Coverage, risk, page metrics

Where Do These Concepts Connect?

Project
├── Test Cases → Test Plans → Executions (via Personas)
├── Personas (encrypted credentials)
├── Discovery Sessions → (generate test cases and docs)
├── Exploratory Sessions → Knowledge Maps
└── Compliance Audits (automated framework checks)

Test Cases are grouped into Test Plans and run via Executions
Personas inject credentials into Executions for auth testing
Discovery finds pages and generates Test Cases and Knowledge docs
Exploratory sessions accumulate Knowledge Maps over time
Compliance Audits run as standalone checks across projects

Ready to Learn More?

Features — Deep dive into each feature
Getting Started — Create your first project
AI Agents — Learn about intelligent test automation

Overview