A test scenario generation framework that finitely enumerates test scenarios through domain definition and relation patterns.
Inspired by routing path computation in network algorithms,
the core idea is: "test scenarios can be derived by SQL-style enumeration."
Test scenario generation is work that tends to depend on individual experience and intuition.
This framework formalizes that process into a finite, reproducible pipeline:
domain definition → boolean operation → pattern enumeration.
What distinguishes it from conventional decision tables or CPP (cause-effect graphs)
is that input value domains are structured through a network routing design analogy,
and it includes an explicitly defined operational model that separates the roles of human, AI, and program.
The core ideas behind the Mirror Framework.
- The claim that test scenarios can be finitely enumerated from domain and relation patterns
- How test data is generated through boolean operations
- How YAML conversion enables structured input to AI
- A role model that clearly separates human, AI, and programmatic processing
A systematic classification of input values required for test design.
- Numeric types (integers, decimals, dates, boundary values, signed numbers)
- String types (full-width, half-width, control characters, special characters)
- HTML encoding and injection-aware classifications
- Assigning boolean flags to each pattern enables AI-driven enumeration
A methodology for extending the Mirror Framework — strong in unit testing —
into integration and system testing.
- Screen transition computation inspired by optical network path design (airport EPS routing)
- Full path enumeration using adjacency tables + LEFT JOIN
- Automatic test phase classification via path cost (0=unit, 1~n=integration, max=system)
- Includes a sample adjacency table built from an actual design document
The schemas/ directory contains YAML schema definitions of the structures described in each document.
These serve as input formats for AI-driven test scenario generation.
- playwright-framework-guide
Design philosophy for Playwright E2E test automation — the intended implementation target of this framework.
This repository handles design philosophy and documentation rather than code,
so the Creative Commons Attribution-ShareAlike license applies.
When citing, adapting, or redistributing, please credit the original author
and apply the same license.