Skip to content

mmaideveloper/aiarchitecttoolkit / skills/architecture-change

architecture-change

Unverified

Coordinate a traceable architecture change from an idea or stakeholder use case through Business Decision Records, Architecture Design Documents, C4 views, Architecture Decision Records, implementation-ready tasks, and post-implementation conformance review. Use when starting, governing, assessing, or continuing a significant architecture change from any artifact reference.

Architecture Change

Coordinate the lifecycle and use focused skills to author artifacts.

Workflow

  1. Read the target repository's root AGENTS.md. When project_name is supplied, load the exact profiles/<project_name>/profile.yaml, its configured project AGENTS.md, and document instructions from the toolkit before authoring artifacts; reject unknown projects. Also read architecture/toolkit-profile.yaml when present, repository sources, existing artifacts, and stakeholder evidence. The target repository's root instructions remain authoritative.
  2. Resolve the starting point and change slug. Do not recreate an artifact already supplied by reference.
  3. Use $idea-task only when the problem or intended outcome is still ambiguous.
  4. Use $generate-bdr before or alongside use-case discovery when an authoritative business choice is required. An accepted BDR may authorize or constrain multiple use cases.
  5. Use $generate-use-case; require Reviewed before architecture baselining.
  6. Use $generate-add from reviewed upstream artifacts and source evidence.
  7. Use $generate-c4 only for views that answer concrete stakeholder questions.
  8. Use $generate-adr once per independently changeable decision.
  9. Reconcile the ADD, BDR impacts, and target-state C4 views after ADR acceptance.
  10. Use $prepare-task to create one or more implementation-ready task specifications. Do not mark a task ready while a blocking decision or compliance question remains open.
  11. After implementation, use $review-architecture-conformance; use $security-review when required by the risk or project profile, and $azure-review when Azure is in scope. Resolve or explicitly govern deviations and findings.
  12. Before an AI solution release or formal gate, use $ai-solution-final-review to synthesize current evidence and recommend readiness without inferring approval.

Read references/artifact-contract.md when creating, linking, or changing lifecycle state for any artifact.

Gates

  • Label material claims Confirmed, Assumption, To verify, or Unknown.
  • Apply the project profile and applicable privacy, AI, security, safety, accessibility, and sector rules. Stop on unresolved conflicts.
  • Never put regulated records, credentials, secrets, or sensitive operational details in artifacts.
  • Never infer approval or authority.
  • Architecture work does not authorize implementation or external task creation unless the user requests it.

Completion

Report artifact paths and states, traceability gaps, accepted versus proposed decisions, specialist-review status, governance outcome, task or release readiness, and the next responsible action.

brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 318kUpdated Claude CodeCodexUnverifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 196kUpdated Claude CodeCodexVerifiedInstall