Creates, reviews, updates, and manages Architecture Decision Records (ADRs) — the institutional memory of technical decision-making. Use this skill whenever the user wants to: create an ADR for a technology or architecture decision, document why a specific technology was chosen, record a technical decision before implementing it, update or supersede an existing ADR, review a proposed decision for completeness, check whether a decision warrants an ADR, maintain the ADR index, or understand why a past decision was made. Also trigger when the user asks "why are we using X", "who decided this", "was this decision documented", "log a design decision", "decision history", "record a technical decision", or "document this choice".
This listing is imported from SkillsMP metadata and should be treated as untrusted until upstream source review is completed.
Install skill "architecture-decision-records" with this command: npx skills add s3nex-com/skillsmp-s3nex-com-s3nex-com-architecture-decision-records
This source entry does not include full markdown content beyond metadata.
This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.
Related by shared tags or category signals.
accessibility, a11y, WCAG, screen reader, keyboard navigation, color contrast, inclusive design, EU Accessibility Act, ADA compliance, focus management, aria labels, accessible components, ARIA, tab order, focus trap, skip link, axe-core, pa11y, contrast ratio, accessible forms, landmark regions
Activate when verifying that a service implementation actually matches its API contract, running contract tests between consumer and provider services, detecting contract drift between what the spec says and what is deployed, setting up Pact or schema-registry based contract verification in CI, investigating a production incident caused by a contract violation, comparing two spec versions to identify breaking changes, validating partner company deliverables against the agreed OpenAPI spec, or enforcing that no spec changes are deployed without going through the change control process. Use this when something is broken at an integration boundary and you need to determine whether it is a contract violation or an implementation bug.
Activate when setting up or running architecture fitness functions, enforcing import boundaries in CI, checking module layer boundaries automatically, tracking dependency budget against an approved list, detecting circular imports, flagging dead or abandoned modules, preventing architecture drift between PRs, enforcing architecture compliance in the build pipeline, or adding architecture CI checks to a project. Distinct from periodic human architecture reviews — fitness functions run automatically on every PR with no human required.
Activate when adding caching to a service, debugging cache-related bugs, configuring a CDN (Cloudflare, Fastly, CloudFront), designing cache invalidation, investigating low cache hit rate, diagnosing a cache stampede, picking TTLs, introducing Redis or Memcached, designing edge caching for static assets or API responses, choosing between cache-aside, read-through, write-through, write-behind, or refresh-ahead patterns, reviewing cache coherency in a distributed system, or deciding whether caching is the right answer at all. Trigger phrases: "add caching", "cache invalidation", "CDN configuration", "cache hit rate", "cache stampede", "cache strategy", "Redis caching", "Cloudflare config".