skills

Use when auditing the overall architecture for completeness, consistency, principle compliance, derivation chain integrity, and proposing/validating adjustments. This skill unifies the meta-verification layer and replaces architecture-self-auditor, derivation-chain-validator, adjustment-proposer, and adjustment-validator.

Use when designing system architecture, recording decisions, selecting patterns and tech stack, analyzing data flow, designing interface contracts, and applying the Strangler pattern. This skill unifies D2 Architecture Derivation Domain and replaces architecture-decision-recorder, architecture-pattern-selector, tech-stack-selector, data-flow-analyzer, interface-contract-designer, and strangler-pattern-suite.

Use when generating implementation code from interface contracts for backend (Go) and frontend (Vue/React) in any language. This skill unifies D5 Implementation Derivation Domain (L4.5→L5) and replaces contract-driven-code-generator, backend-code-generator, frontend-code-generator, and generic-code-generator.

Use when validating code and specifications against all 15 Aether constitutional principles (P0-P14), detecting principle conflicts, resolving via dynamic weighting, and enforcing mechanized constraints. This skill unifies the constitution enforcement layer and replaces constitution-validator, principle-consistency-checker, and constraint-check-runner.

Use when orchestrating complex deployments (canary, blue-green, rolling), managing releases, handling rollbacks, and enforcing change management. This skill unifies deployment operations and replaces deployment-orchestrator, rollback-manager, release-manager, and change-management.

Use when evaluating task determinism, dispatching to traditional code generators vs AI generation, enforcing contract consistency gates, and routing for confidence-based human review. This skill unifies D4.5 Generation Dispatch Domain and replaces deterministic-evaluator, code-generator-dispatcher, contract-consistency-gate, and confidence-based-reviewer.

Use when initializing a complete Go + Vue + Quasar fullstack project with production-ready structure, scaffolding, Makefile, Docker, CI/CD, and bilingual documentation. This skill unifies the fullstack development toolchain and replaces go-backend-scaffolder, go-cli-builder, vue-quasar-scaffolder, fullstack-project-setup, go-vue-fullstack-workflow, makefile-backend-generator, and readme-license-generator.

Use when managing production incidents with ITIL-aligned processes: detection, response, resolution, root cause analysis, and knowledge base updates. This skill unifies incident and problem management.

Use when tracking AI collaboration effectiveness (token savings, generation quality, review reduction) and system quality (completeness, coverage, correctness) for dual-track recursive optimization. This skill unifies the metrics layer and replaces ai-effectiveness-tracker and system-quality-tracker.

Use when optimizing skills, prompts, and workflows based on execution results and feedback, with convergence guarantees (Lipschitz learning rate). This skill unifies P13 Recursive-Self-Optimization and replaces recursive-optimizer, convergence-checker, improvement-budget-allocator, and tech-debt-quantifier.

Use when collecting business requirements, decomposing goals into functional requirements, classifying them into 13 categories, mapping to business value via BMAD, defining metrics, and establishing traceability. This skill unifies L1→L2 derivation (D1 Value Derivation Domain) and replaces business-requirements-collector, business-value-mapper, requirement-classifier, value-decomposer, metrics-definer, and requirements-to-code-docs.

Use when extracting specifications, architecture, and behavior from existing legacy systems for L1 derivation in brownfield projects. Implements characterization tests, static + dynamic analysis, and progressive documentation.

Use when detecting the current development scenario, recommending the appropriate mode (norm-driven, emergent exploration, intent-driven), configuring principle strengths per P0, and managing scenario-specific exemptions. This skill unifies D6 Scenario Adaptation Domain and replaces scenario-detector, scenario-mode-recommender, mode-selector, poc-exemption-manager, dual-track-validator, ai-native-build-workflow, standard-mode-workflow, federal-constitution-manager, and harness-reverse-construction-workflow.

Use when managing the skill library lifecycle (creation, validation, assetization, application, feedback, optimization, retirement), identifying skill gaps, recommending skills for tasks, and packaging skills for distribution. This skill unifies P14 Knowledge-Engine-Skill-Assetization and replaces skill-lifecycle-manager, skill-optimizer, skill-recommender, skill-gap-analyzer, skill-packaging-tool, and prompt-template-manager.

Use when parsing requirements into structured specifications, designing use cases, writing Gherkin BDD scenarios, generating ATDD acceptance tests, validating scenario completeness, analyzing NFRs, and managing spec versions. This skill unifies D3 Specification Derivation Domain and replaces spec-parser, usecase-designer, bdd-scenario-writer, atdd-acceptance-test-generator, spec-to-scenario, scenario-completeness-checker, nfr-analyzer, and spec-version-manager.

Use when deriving test stratification plans from GWT scenarios, generating tests at all pyramid layers (unit/integration/contract/E2E/system/SIT), validating test proportions, and designing chaos experiments. This skill unifies D4 Contract Derivation Domain and replaces test-pyramid-deriver, test-pyramid-validator, e2e-test-generator, integration-test-generator, contract-test-generator, system-test-generator, unit-test-generator, sit-scenario-generator, chaos-test-designer, and non-functional-gate-checker.

Use when performing cross-domain skepticism verification covering correctness, completeness, consistency, and boundedness. This skill unifies the skeptic verification layer and replaces correctness-checker, completeness-checker, consistency-checker, and boundedness-checker.

Use when managing production emergencies with two-phase process: immediate response (minutes) followed by post-incident normalization (24h). Implements incident command, on-call escalation, and battle rhythm.

Use when layering business intent into three levels (business intent, functional intent, implementation intent) for intent-driven development. Implements P11 Intent-Hierarchization Principle.

Use when orchestrating end-to-end software lifecycle with Aether.go methodology fusion framework across all eight stages

Use when refactoring implementation code to improve design while maintaining test coverage, supporting P9 recursive self-optimization

Use when challenging boundary rationality of test-pyramid-deriver outputs with META-005 protocol, requiring explicitized argumentation before accepting derivation results

Use when generating structured reverse clarification questions at critical requirement semantic nodes to prevent perfect derivation but wrong software

Use when executing the complete TDD red-green-refactor cycle as an atomic skill, bridging L4 test specifications to L5 implementation

Use when managing project context, conversation history, and information retrieval for efficient AI assistance. Implements P7 Context-First Principle.