Menu
Skills in this repository
SkillsMP has collected 98 skills from navikt/grillmester. Open a skill to review its source and details.
navikt/grillmesterShowing 40 of 98 collected skills.
Writes or edits clear Norwegian prose without changing its technical meaning. Use for user-facing text, documentation, README text, ADR prose, pull-request descriptions or requests to remove anglicisms and AI-like phrasing.
Source text: Norwegian Bokmål
Implement or review Kotlin and Ktor backend work in Nav. Use for routes, plugins, startup, modules, dependency injection, configuration, structured logging and MDC, outgoing HttpClient, graceful shutdown, or Ktor integration with authentication, Kafka, and…
Integrate or review @navikt/lumi-survey in a Nav frontend and its backend-for-frontend. Use for survey selection and configuration, package setup, widget rendering, submission transport, TokenX or Azure/Entra token exchange, Nais accessPolicy, privacy-safe…
Create, change, or review a Nais Application or Naisjob manifest. Use for ingress, resources, probes, accessPolicy, Azure or TokenX, Kafka, GCP Postgres, observability, scaling, CPU throttling, OOM behavior, graceful shutdown, or scheduled batch jobs.
Diagnoses failures in deployed Nais workloads using runtime evidence from Kubernetes, deployment status, metrics, logs and traces. Use for pod startup failures, deploy failures, 401/403 responses, message-consumer lag, database connectivity, latency, restarts…
Designs or improves metrics, structured logging, tracing, health checks, dashboards and alerts for deployed services. Use for instrumentation, correlation, PromQL/LogQL, operational dashboards, SLO signals or Nais observability setup.
Helps formulate and review outcome-oriented goals for public-sector product work. Use for OKRs, objectives and key results, planning cycles, goal reviews, baselines, measurement plans or converting delivery lists into measurable outcomes.
Review or design PostgreSQL use in a Nav backend. Use for a DataSource, HikariCP pool, Cloud SQL capacity, connection errors, schema or query review, Flyway migration, indexes, N+1 behavior, shared databases, destructive changes, or database technology choice.
Supports public-sector product discovery, opportunity trees, assumption testing, user interviews, product-risk reflection and product-lead competency development. Use for open problem spaces, initiative risk, discovery planning or self-evaluation.
Build a bounded, runnable throwaway prototype to resolve an uncertain behavior or interface before committing to a production design. Use for data models, state machines, API or error contracts, event flows, retries, and other questions that benefit from…
Creates or updates repository README documentation from verified implementation and operational evidence. Use when a README is missing, stale or needs accurate coverage of purpose, setup, interfaces, data flows, deployment, observability or ownership.
Self-review the complete task-scoped diff before independent review or delivery. Use after implementation, before a pull request, or when asked to inspect uncommitted work, a branch against a fixed point, or changes since a named revision.
Runs vertical red-green-refactor cycles through a repository's discovered public interfaces and test tooling. Use when test-first or red-green-refactor is explicitly requested, or when a bug fix needs a reproduction test before implementation; use…
Builds evidence-based team status, goal progress and prioritisation summaries from explicitly confirmed sources. Use for weekly status, planning-period reviews, cross-repository work overviews or prioritisation; GitHub Projects is supported but never assumed.
Use only after the user explicitly selects To Issues to break an approved plan or specification into independently useful GitHub issues with native blocking edges.
Use only after the user explicitly selects To Spec to turn an already resolved conversation or design into a concise engineering specification; do not interview or reopen decisions.
Assesses incoming issues and pull requests, verifies claims, identifies missing context and produces work-ready briefs. Use for inbox review, bug-report assessment, prioritisation, readiness decisions or a specific issue/PR that needs triage.
Use only after the user explicitly selects Wayfinder to plan work too large for one agent session as a shared map of decision tickets, resolved one at a time until the route is clear.
Designs outcome-driven workshops, meetings, retrospectives, foundation sprints and team-health follow-up with facilitation plans and tool-neutral board structures. Use when a group needs exploration, alignment, a decision or concrete follow-up.
Review consequential architecture proposals using repository evidence, boundaries, quality attributes, operations, alternatives, migration, and reversibility. Use for new services, cross-team contracts, platform choices, migrations, or hard-to-reverse design…
Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.
Finds deepening opportunities that improve module depth, locality, leverage, test seams, and navigability. Use when the user asks to improve architecture, consolidate tightly coupled or shallow modules, find refactoring candidates, or explain why code is…
Build a bounded, runnable throwaway prototype to resolve an uncertain behavior or interface before committing to a production design. Use for data models, state machines, API or error contracts, event flows, retries, and other questions that benefit from…
Explores, designs and evaluates ethically responsible behavioural interventions in public services. Use for nudging, behavioural design, reminders, defaults, friction or sludge, action-triggering copy, and experiments intended to influence user behaviour.
Helps formulate and review outcome-oriented goals for public-sector product work. Use for OKRs, objectives and key results, planning cycles, goal reviews, baselines, measurement plans or converting delivery lists into measurable outcomes.
Supports public-sector product discovery, opportunity trees, assumption testing, user interviews, product-risk reflection and product-lead competency development. Use for open problem spaces, initiative risk, discovery planning or self-evaluation.
Builds evidence-based team status, goal progress and prioritisation summaries from explicitly confirmed sources. Use for weekly status, planning-period reviews, cross-repository work overviews or prioritisation; GitHub Projects is supported but never assumed.
Designs outcome-driven workshops, meetings, retrospectives, foundation sprints and team-health follow-up with facilitation plans and tool-neutral board structures. Use when a group needs exploration, alignment, a decision or concrete follow-up.
Integrate or review @navikt/lumi-survey in a Nav frontend and its backend-for-frontend. Use for survey selection and configuration, package setup, widget rendering, submission transport, TokenX or Azure/Entra token exchange, Nais accessPolicy, privacy-safe…
Aksel-designsystem og Nav-frontendmønstre — komponentvalg, layout, spacing/tokens, skjema og responsiv UI med @navikt/ds-react. Brukes via /grillmester-aksel-design ved nye komponenter, layout-endringer eller styling-valg, og før rå HTML/CSS eller custom…
Source text: Norwegian Bokmål
UU/WCAG-review for Nav-frontend — tastaturflyt, skjermleser, axe, kontrast, fokus, feilmeldinger, skjema og Aksel-bruk. Brukes via /grillmester-accessibility-review før PR eller release.
Source text: Norwegian Bokmål
Design or change a backend API contract in Nav. Use for new endpoints, versioning, breaking changes, deprecation, consumer discovery, accessPolicy implications, compatibility gates, or API catalogue publication.
Set up, change, or troubleshoot authentication and authorization in a Nav backend. Use for incoming JWT validation, TokenX OBO, Azure or Entra ID M2M, Texas, Maskinporten, Wonderwall, accessPolicy, protected endpoints, 401 or 403 errors, and issuer or…
Create, revise, review, or diagnose a GitHub Copilot CLI skill with a predictable authoring and forward-testing workflow. Use when a user asks to create or improve a skill, investigate why one does not trigger, or validate skill behavior.
Builds a tight reproduction loop, minimizes the symptom, tests ranked hypotheses, and locks the fix down with regression evidence. Use when something throws, fails, hangs, flakes, regresses in performance, or only fails at runtime.
Audits whether the current repository is ready for effective Grillmester use across Copilot CLI, the Copilot app, cloud agent, and Copilot code review. Use only when the user explicitly asks to check, diagnose, or understand the repository's Grillmester…
Designs full-system tests that boot the application and prove an observable user or system flow across real boundaries. Use when a test must exercise the assembled app, an HTTP or message entry point, and its externally visible outcome; use…
Design-til-kode-arbeidsflyt — Figma-lenker, Code Connect og mapping fra design til Aksel-komponenter, tokens og responsiv layout. Brukes via /grillmester-figma-workflow når oppgaven har Figma-lenke eller design-to-code.
Source text: Norwegian Bokmål
A relentless interview to sharpen a plan or design.
A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.