From architecture-intelligence
Use whenever code work has structural consequences: project architecture, module boundaries, dependency direction, runtime topology, ownership, ADRs, fitness functions, or architecture refactoring.
How this skill is triggered — by the user, by Claude, or both
Slash command
/architecture-intelligence:architecture-intelligenceThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use for architecture quality in code. Trigger from task context and code
Use for architecture quality in code. Trigger from task context and code evidence: designing a new project or feature, choosing structure, changing boundaries, moving behavior between layers, introducing runtime/integration paths, protecting quality attributes, or reviewing code whose shape affects changeability, reliability, scalability, security, operability, or testability.
Before judging:
If context is thin, inspect source first or ask one decision-critical question. Do not invent scale, teams, boundaries, compliance, or domain facts.
From the plugin root:
python3 scripts/architecture_probe.py <repo-path> --json
python3 scripts/architecture_probe.py <repo-path> --json --git-history
python3 scripts/architecture_probe.py <repo-path> --json --policy <policy.json>
Probe output is an evidence map, not a verdict. It includes:
Use the smallest focused skill set:
codebase-architecture-audit: current architecture, risks, scorecards.architecture-conformance: intended vs observed rules, drift, erosion.architecture-ownership-topology: CODEOWNERS/OWNERS, ownership coverage, cross-owned coordination risk, governance paths.architecture-runtime-topology: deployment, runtime calls, observability, resilience, operability.architecture-decisions: ADR creation/review, tradeoffs, owner, revisit trigger.architecture-fitness-functions: executable guardrails, CI checks, review gates.architecture-refactoring-strategy: incremental migration, blast radius, rollback.When the current work will create, preserve, or materially change architecture, do not stop at the router. Use this minimum chain:
codebase-architecture-audit for observed architecture and risks.architecture-refactoring-strategy for target boundary, slices, rollback.architecture-fitness-functions for executable guardrails.Add architecture-conformance when an architecture doc, ADR, policy, or
explicit constraint exists. Add architecture-runtime-topology when CLI, app
refresh, background work, deployment, integration, or operability paths change.
Add architecture-ownership-topology when owner files, package ownership, or
cross-owned dependencies affect the change.
Use adjacent plugins for UI/UX, game design, Kotlin/Tauri specifics, security threat modeling, or non-architecture cleanup.
Before editing, capture dirty-tree state, the target boundary, and a pre-refactor probe when the repo can be scanned. After editing, provide tests, docs/ADR updates, runtime smoke or an explicit skip reason, and a post-probe comparison when structural metrics informed the change. End with a skill-usage self-check: skills used, specialized skills skipped, and the reason each skip was safe.
For durable handoff, emit architecture_intelligence.refactor_report.v1.
Use references/contracts.md for:
architecture_intelligence.audit.v1architecture_intelligence.conformance.v1architecture_intelligence.decision.v1architecture_intelligence.fitness_plan.v1architecture_intelligence.refactor_report.v1architecture_intelligence.runtime_topology.v1architecture_intelligence.structure_metrics.v1architecture_intelligence.ownership_topology.v1Do not approve high-risk architecture changes without validation. Do not prescribe microservices, event sourcing, DDD, clean/hexagonal architecture, or new tooling by default. Optimize codebase changeability and runtime qualities, not diagram polish.
npx claudepluginhub xopoko/plug-n-skills --plugin architecture-intelligenceProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.