Release: v0.16.14

Fact Disposition Framework

Existing-behavior facts surfaced by codebase analysis are now explicitly traced through to Design Docs, reviewed for completeness, and checked for cross-layer consistency.

codebase-analyzer

• focusAreas repurposed as disposition targets with fact_id, evidence, and factsToAddress fields
• New Step 4.5: disposition target enumeration (input fields, call sites, branching cases, data shapes, error paths, external dependencies, operational cases)
• Cardinality guidance: 5-15 entries per analysis, grouped by coherent fact unit

technical-designer / technical-designer-frontend

• Added "Fact Disposition" required section: one table row per focusArea with disposition (preserve / transform / remove / out-of-scope), rationale, and evidence
• Added "Prior-Layer Verification" input parameter for fullstack flows

document-reviewer

• Added codebase_analysis input parameter
• Gate 0: structural check for Fact Disposition Table presence and focusArea coverage
• Gate 1: disposition value validation, rationale completeness, evidence carry-through

design-sync

• Added "Disposition conflict" detection type: same fact_id across Design Docs with different disposition values → critical severity

Design Doc template

• Added "Fact Disposition Table" section (Fact ID, Focus Area, Disposition, Rationale, Evidence)

Fullstack Flow Improvements

monorepo-flow

• Backend Design Doc is now verified by code-verifier before frontend Design Doc authoring begins
• prior_layer_verification result explicitly passed to the frontend designer
• Layer Context templates updated; Task Cycle section deduplicated

Orchestration Improvements

subagents-orchestration-guide

• Handoff Contracts formalized as HC-01 through HC-06
• Basic Flow consolidated from numbered step lists into compact table + rules

Release: v0.16.13

Quality Assurance Mechanism Pipeline

codebase-analyzer

• Added qualityAssurance output section that discovers linters, validators, and domain-specific constraints covering the affected files

technical-designer

• Extended Standards Identification Gate with quality assurance mechanism adoption decisions (adopted vs noted with reason)

Design Doc / plan / task templates

• Added "Quality Assurance Mechanisms" section to Design Doc template with coverage scope
• Added QA Mechanisms table with Covered Files column to plan template
• Added QA Mechanisms section to task template

work-planner / task-decomposer

• work-planner extracts adopted QA mechanisms from Design Doc and propagates to plan header
• task-decomposer uses file coverage overlap to deterministically match mechanisms to tasks

quality-fixer / quality-fixer-frontend

• Added optional task_file input parameter for supplementary quality gate hints
• Added taskFileMechanisms field to approved and blocked responses for auditability
• All orchestration recipes now always pass task file path to quality-fixer

Reference Representativeness

coding-principles

• Added "Reference Representativeness" section: verify patterns and dependencies from nearby code are representative across the repository before adoption

ai-development-guide

• Extended Pattern 5 (Insufficient Existing Code Investigation) with reference representativeness check

task-executor

• Added Reference Representativeness check during implementation with repository-wide verification
• Added dependency_version_uncertain escalation type when repo state alone is insufficient to determine the appropriate version

Release: v0.16.12

Diagnose Flow: Failure Point Model

investigator / verifier / solver

• Replaced hypothesis-based reasoning with a failure point model in the diagnose workflow
• Investigator now identifies concrete failure points instead of generating abstract hypotheses
• Verifier validates failure points against evidence rather than ranking hypotheses
• Solver derives solutions from verified failure points with clearer causal chains

E2E Test Generation Fix

acceptance-test-generator

• Fixed ROI calculation that made E2E test selection mathematically impossible (max score ~3.16 against threshold of 50)
• Simplified ROI to raw Value Score (0-120) for same-type candidate ranking
• Replaced blanket push-down rule ("already integration-tested → no E2E") with journey-aware retention
• Added reserved E2E slot for user-facing multi-step journeys, emitted regardless of ROI threshold
• Added structured generation report with nullable generatedFiles.e2e and e2eAbsenceReason

integration-e2e-testing skill

• Added environment-agnostic multi-step user journey definition (Web, mobile, CLI, API)
• Added user-facing vs service-internal classification for reserved slot eligibility
• Added ROI calculation examples showing selection outcomes

work-planner

• Added E2E Gap Check in common planning path (fires for both Strategy A and B)
• Gap Check respects e2eAbsenceReason to skip warning on intentional E2E absence

Downstream contract updates

• Updated orchestration guide to treat nullable E2E as valid (not an error)
• Updated recipe-plan, recipe-front-plan, recipe-implement, recipe-fullstack-implement to propagate e2eAbsenceReason through the handoff chain

Release: v0.16.11

Incomplete Implementation Detection Gate

quality-fixer / quality-fixer-frontend

• Added Step 1 "Incomplete Implementation Check" as a blocking gate before quality checks
• Reviews uncommitted diff for stub patterns (TODO/FIXME markers, placeholder returns, empty method bodies, deferred-implementation comments)
• Includes positive and negative detection examples to reduce false positives
• Returns new stub_detected status with structured response listing incomplete implementations
• Quality checks are skipped entirely when stubs are found — verifying unfinished code is meaningless

Orchestration & Recipes

• Orchestrator routes stub_detected back to task-executor for completion, then re-runs quality-fixer
• All consuming recipes updated with stub_detected branch handling: recipe-build, recipe-front-build, recipe-fullstack-build, recipe-implement, recipe-fullstack-implement, recipe-add-integration-tests
• Unified approved status references across all orchestration and recipe files
• Normalized step heading levels for consistent LLM structural parsing

v0.16.10

Cross-Document Consistency Verification

design-sync

• Redesigned as a high-recall candidate generator with Match Basis Rules
• Conflicts now carry confidence (high/medium) and match_basis metadata
• Output separated into [CONFIRMED_CONFLICTS] and [CANDIDATE_CONFLICTS] sections
• Added path identifiers to extraction targets with structured extraction output format
• Cross-section matching enabled (section name differences no longer block detection)
• Medium confidence match bases (same_endpoint_role, same_integration_role, same_ac_slot) require structural evidence in reason field

code-verifier

• Added literal identifier referential integrity checks (paths, endpoints, type names, config keys verified against codebase definitions)
• Single authoritative definition now sufficient for high-confidence identifier existence verification

technical-designer / technical-designer-frontend

• Added Dependency Inventory for update mode, requiring identifier verification against codebase and Accepted ADRs before document modification

recipe-update-doc

• Added literal identifier referential integrity focus hint to code-verifier invocation

Release: v0.16.9

UI Spec Design Token Concreteness

Raised the specificity of UI Spec template design tokens, inspired by the DESIGN.md (Stitch format) approach of concrete token-level values over vague descriptions.

ui-spec-template

• Restructured Design Tokens from a flat table into semantic sub-sections: Color Roles, Typography Hierarchy, Spacing Scale, Elevation (Depth), Border Radius Scale
• Expanded Environment Constraints with Responsive Behavior table specifying concrete key changes per breakpoint
• Improved State x Display Matrix examples with existing component references

AC ID Traceability

Added sequential AC IDs to PRD template for consistent traceability across the PRD → UI Spec → Design Doc pipeline.

prd-template

• Acceptance criteria now use sequential IDs (AC-001, AC-002, ...) instead of unlabeled AC: entries

documentation-criteria

• PRD definition explicitly requires AC IDs for downstream traceability
• AC ID format aligned across PRD, UI Spec, and Design Doc templates

Release: v0.16.8

Design-to-Plan Traceability

Added a traceability mechanism to ensure Design Doc technical requirements are not silently dropped during work plan generation.

work-planner

• New Step 5: scan all DD sections using category-based extraction (implementation target, connection/switching/registration, contract change, verification requirement, prerequisite work) and map each item to a covering task
• Design-to-Plan Traceability table required in every work plan
• Gaps without justification are errors; justified gaps require user confirmation before plan approval

plan-template

• Added Design-to-Plan Traceability section with category-based mapping table and gap status tracking

design-template

• Added Interface Change Matrix section (existing/new interface, conversion required, compatibility method)
• Added Verification Method column to Integration Points List

subagents-orchestration-guide

• Expanded technical-designer → work-planner handoff with category-based extraction requirements and gap escalation rules

Release: v0.16.7

Deep codebase analysis and output comparison verification

Strengthened the analysis-to-design pipeline to prevent design documents from being based on incomplete understanding of existing code.

codebase-analyzer

• Extract all methods (public and private/internal) instead of public interfaces only
• Trace full call chains within modules instead of single-level dependency tracing
• New dataTransformationPipelines output: step-by-step input/output mapping with external resource lookups (master tables, config, constants)

technical-designer

• Removed method listing limit — list every public method with full signatures
• Added output comparison requirement for designs that replace or modify existing behavior
• Connected dataTransformationPipelines to Verification Strategy input contract

document-reviewer

• Added output comparison check: missing output comparison for behavior-replacing changes is flagged as critical
• Pipeline step coverage verified when dataTransformationPipelines are referenced

design-template

• Added Output Comparison section under Verification Strategy (comparison input, expected output fields, diff method, transformation pipeline coverage)

subagents-orchestration-guide

• Pass dataTransformationPipelines in codebase-analyzer to technical-designer bridge

Release: v0.16.6

Frontend recipe alignment with backend practices

Aligned all four frontend recipe skills with backend conventions to improve LLM execution consistency.

• Unified Orchestrator Definition to Execution Protocol format with subagents-orchestration-guide references
• Added design-sync quality gate to recipe-front-design with correct stop point ordering
• Added Workflow Overview and Completion Criteria to recipe-front-design
• Made test skeleton generation user-confirmed in recipe-front-plan, matching backend behavior
• Removed redundant sections from recipe-front-build (Execution Method, Sub-agent Invocation Method, Structured Response Specification)
• Replaced rule-advisor with documentation-criteria in recipe-front-review fix flow
• Fixed re-validation prompt to include Design Doc path and implementation file list
• Clarified guide references to avoid "exactly" conflicts with recipe-level simplifications

Release: v0.16.5

Stricter Design Doc compliance verification

Implementation accuracy now has two layers of defense: a strengthened code-reviewer for on-demand reviews, and an automated post-implementation verification gate that runs code-verifier and security-reviewer in parallel before completion.

code-reviewer

• Identifier verification with exact string matching against Design Doc
• Multi-source evidence collection with confidence levels (high/medium/low)
• Finding classification (dd_violation/maintainability/reliability/coverage_gap) with mandatory rationale
• Quality thresholds delegated to coding-principles skill instead of hardcoded values
• Output self-check requiring tool result citations

Post-implementation verification gate

• code-verifier added alongside security-reviewer, running in parallel after all tasks complete
• Diff-scoped verification focused on changed files
• Pass/fail criteria and re-run rules defined per verifier

Orchestration guide

• Reduced redundancy by removing sections that restated upstream definitions