Skip to content

CHANGELOG.md

Latest commit

 

History

History
129 lines (112 loc) · 9.7 KB

CHANGELOG.md

File metadata and controls

129 lines (112 loc) · 9.7 KB

Changelog

All notable changes to the V-Model Extension Pack are documented here.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.4.0] — 2026-02-22

Added

  • module-design command — DO-178C/ISO 26262-compliant low-level module designs with four mandatory views (Algorithmic/Logic, State Machine, Internal Data Structures, Error Handling & Return Codes)
  • unit-test command — ISO 29119-4 white-box unit test plans with five named techniques (Statement & Branch Coverage, Boundary Value Analysis, Equivalence Partitioning, State Transition Testing, Strict Isolation) and Dependency & Mock Registries
  • validate-module-coverage.sh / validate-module-coverage.ps1 — Deterministic ARCH→MOD→UTP→UTS bidirectional coverage validation with EXTERNAL and CROSS-CUTTING module support
  • Matrix D (Unit Verification) in traceability matrix — ARCH → MOD → UTP → UTS with parent ARCH annotations
  • --require-module-design, --require-unit-test flags for setup-v-model (bash + PowerShell)
  • Module design and unit test fixtures across all scenario directories (minimal, complex, gaps, empty, golden)
  • Module-level validators (validate_module_design(), validate_unit_test() in template_validator.py; module_validators.py)
  • MOD-NNN, UTP-NNN-X, UTS-NNN-X# ID patterns in id_validator.py
  • EXTERNAL and DERIVED MODULE tags for third-party and emergent module designs
  • Pester test suite: Validate-Module-Coverage.Tests.ps1 (16 tests)
  • Module design and unit test LLM-as-judge quality metrics (completeness, logic quality, data structure precision, coverage quality, technique appropriateness, isolation strictness)
  • E2E evaluation tests for module-design and unit-test commands (4 tests each)
  • docs/id-schema-guide.md — Comprehensive guide to the four-tier ID schema, intra-level vs inter-level linking, lifecycle, and end-to-end traceability examples

Changed

  • Extension version bumped from 0.3.0 to 0.4.0
  • setup-v-model.sh/ps1 now detects module-design.md and unit-test.md in AVAILABLE_DOCS; 8 symmetric require flags
  • build-matrix.sh/ps1 extended with Matrix D generation
  • trace.md updated from triple-matrix to quadruple-matrix output (A + B + C + D)
  • Test fixture directories expanded from 6 to 8 V-Model files each (+module-design.md, +unit-test.md)
  • Renamed validate-coverage → validate-requirement-coverage across all scripts, tests, docs, and specs for consistent validate-{design-level}-coverage naming convention
  • Documentation updated for v0.4.0: README (12-step workflow, 9 commands, 4-tier ID schema), CONTRIBUTING, SECURITY, compliance-guide, usage-examples, v-model-config, v-model-overview, product-vision
  • Total commands: 7 → 9; BATS tests: 67 → 91; Pester tests: 67 → 91; Structural evals: 37 → 51; LLM-as-judge evals: 26 → 36; E2E evals: 24 → 32

Fixed

  • BATS test for validate-system-coverage partial mode now correctly expects exit 0 (script was updated in v0.2.0 but test was not)
  • PowerShell validate-system-coverage.ps1 now supports partial mode when system-test.md is absent (parity with bash script)
  • PowerShell validate-system-coverage.ps1 handles empty files via null-coalescing (Get-Content -Raw returns $null for 0-byte files)
  • Minimal module-design fixture now includes typed function signatures and complete type definitions for all pseudocode references

[0.3.0] — 2026-02-21

Added

  • architecture-design command — IEEE 42010/Kruchten 4+1 architecture decomposition with Logical, Process, Interface, and Data Flow views
  • integration-test command — ISO 29119-4 integration testing with Interface Contract, Data Flow, Fault Injection, and Concurrency techniques
  • validate-architecture-coverage.sh / validate-architecture-coverage.ps1 — Deterministic ARCH→ITP→ITS bidirectional coverage validation with CROSS-CUTTING module support
  • Matrix C (Integration Verification) in traceability matrix — SYS → ARCH → ITP → ITS with parent REQ annotations
  • --require-system-design, --require-system-test, --require-architecture-design, --require-integration-test flags for setup-v-model (bash + PowerShell)
  • Architecture and integration test fixtures across all scenario directories (minimal, complex, gaps, empty, golden)
  • Architecture-level validators (architecture_validators.py) and structural/E2E evaluations
  • ARCH-NNN, ITP-NNN-X, ITS-NNN-X# ID patterns in id_validator.py
  • CROSS-CUTTING module tag for infrastructure/utility architecture modules
  • Pester test suite: Validate-Architecture-Coverage.Tests.ps1 (15 tests)
  • Architecture and integration LLM-as-judge quality metrics

Changed

  • Extension version bumped from 0.2.0 to 0.3.0
  • setup-v-model.sh/ps1 now detects architecture-design.md and integration-test.md in AVAILABLE_DOCS; 6 symmetric require flags
  • build-matrix.sh/ps1 extended with Matrix C generation
  • trace.md updated from dual-matrix to triple-matrix output (A + B + C)
  • Test fixture directories consolidated to shared scenario pattern (minimal, complex, gaps, empty) with 6 V-Model files each
  • All Pester test fixture paths updated for consolidated directory structure
  • Documentation updated for v0.3.0: README (9-step workflow, 7 commands, 3-tier ID schema), CONTRIBUTING, SECURITY, product-vision, v-model-config
  • Total commands: 5 → 7; BATS tests: 48 → 67; Pester tests: 48 → 67; Structural evals: 21 → 37; LLM-as-judge evals: 16 → 26

[0.2.0] — 2026-02-20

Added

  • /speckit.v-model.system-design command — Decomposes requirements into IEEE 1016-compliant system components
    • Four mandatory design views: Decomposition, Dependency, Interface, Data Design
    • Many-to-many REQ↔SYS traceability with derived requirements support
    • SYS-NNN ID schema with parent requirement references
  • /speckit.v-model.system-test command — Generates ISO 29119-compliant system test plans
    • Named testing techniques: Interface Contract Testing, Boundary Value Analysis, Fault Injection, Equivalence Partitioning, State Transition Testing
    • Technical BDD scenarios (no user-journey language) with STP-NNN-X / STS-NNN-X# IDs
    • Anti-pattern guard: rejects user-journey phrasing in system-level tests
  • Extended /speckit.v-model.trace command — Dual-matrix traceability output
    • Matrix A: REQ → ATP → SCN (acceptance-level, existing)
    • Matrix B: REQ → SYS → STP → STS (system-level, new)
    • Combined coverage metrics across both matrices
  • System-level golden examples for evaluation:
    • Medical device (CBGMS): IEC 62304 Class C, 5 SYS components, 10 STP test cases
    • Automotive ADAS (AEB): ISO 26262 ASIL-D, 5 SYS components, 11 STP test cases
  • E2E evaluation harness (tests/evals/harness.py) — faithfully simulates spec-kit command invocation via LLM
  • 16 E2E evaluation tests (4 per command: structural + quality for each domain)
  • Structural evaluations in PR CI (26 deterministic tests, no API keys required)
  • Templates for system design and system test output documents
  • Helper scripts for system-level coverage validation (Bash + PowerShell)

Changed

  • Template validators now accept both template-style ("Overview") and golden-fixture-style ("Document Control", "Test Strategy") sections
  • validate-requirement-coverage and build-matrix scripts extended for dual-matrix support
  • Evals workflow updated with E2E job for command invocation testing

[0.1.0] — 2026-02-19

Added

  • Extension scaffold with extension.yml manifest (schema v1.0)
  • /speckit.v-model.requirements command — Generates V-Model Requirements Specification
    • IEEE 29148 / INCOSE 8-criteria quality validation (Unambiguous, Testable, Atomic, Complete, Consistent, Traceable, Feasible, Necessary)
    • Banned words table enforcing measurable, testable language
    • Four requirement categories: Functional (REQ-), Non-Functional (REQ-NF-), Interface (REQ-IF-), Constraint (REQ-CN-)
    • Strict translator constraint for spec.md → REQ-NNN extraction
  • /speckit.v-model.acceptance command — Generates three-tier Acceptance Test Plan
    • Test Cases (ATP-NNN-X) with 4 quality criteria (Traceable, Independent, Repeatable, Clear Expected Result)
    • BDD Scenarios (SCN-NNN-X#) with 4 quality criteria (Declarative, Single Action, Strict Preconditions, Observable Outcomes)
    • Batched generation (5 requirements per batch) to prevent token bloat
    • Deterministic 100% coverage validation gate via helper script
    • Append-only incremental updates with git diff change detection
  • /speckit.v-model.trace command — Builds regulatory-grade Bidirectional Traceability Matrix
    • 4 pillars: Strict Bidirectionality, Orphan & Gap Analysis, Versioning & Baselines, Granular Execution State
    • 3-section output: Coverage Audit, Exception Report, 3D Matrix
    • Deterministic script-based matrix generation (not AI-generated)
  • Output templates for requirements, acceptance plan, and traceability matrix
  • Helper scripts (Bash + PowerShell):
    • setup-v-model — Directory setup and prerequisite checking
    • validate-requirement-coverage — Deterministic REQ→ATP→SCN coverage validation
    • build-matrix — Deterministic traceability matrix builder
    • diff-requirements — Detects changed/added requirements for incremental updates
  • Extension configuration template (config-template.yml)
  • Documentation:
    • v-model-overview.md — V-Model methodology context
    • usage-examples.md — Medical device (IEC 62304) and automotive (ISO 26262) examples
    • compliance-guide.md — Artifact mapping to IEC 62304, ISO 26262, DO-178C, FDA 21 CFR Part 820, IEC 61508
  • after_tasks hook to automatically run traceability matrix after task generation
  • Self-documenting three-tier ID schema: REQ-NNN → ATP-NNN-X → SCN-NNN-X#