Nav Architecture Review logo

Nav Architecture Review

Organization
navikt
nav-architecture-review

Generer Architecture Decision Records (ADR) med flerperspektiv-review tilpasset Nav

Overview

Publishernavikt
Repositorycopilot
Skill namenav-architecture-review
Stars
60
Forks
16
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by navikt on GitHub. Read the source before you install it.

Installation

Install the Nav Architecture Review AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/navikt/copilot/tree/main/skills/nav-architecture-review
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/navikt/copilot.git /tmp/copilot
mkdir -p .claude/skills
cp -r /tmp/copilot/skills/nav-architecture-review .claude/skills/nav-architecture-review
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Nav Architecture Review in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Nav Architecture Review on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Nav Architecture Review is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Architecture Review — ADR Generator

Generate an Architecture Decision Record (ADR) with review from five perspectives: architecture, security, platform, migration, and user experience. Follow Nav's Architecture Advice Process.

Workflow

  1. Forstå konteksten — hva er endringen og hvorfor?
  2. Review the five perspectives — architecture, security, platform, migration, and user experience; state when a perspective does not apply.
  3. Identifiser alternativer — minst to alternativer + valgt løsning
  4. Generer ADR — formelt dokument med Nav-spesifikke vurderinger
  5. Liste aksjonspunkter — hva må gjøres for å realisere beslutningen

Steg 1: Kontekst

Still disse spørsmålene:

  • Hva er endringen? (én setning)
  • Hvorfor gjøres den? (forretningsbehov, teknisk gjeld, regulatorisk)
  • Hva er konsekvensen av å ikke gjøre noe?
  • Hvilke team påvirkes?

Steg 2: Fler-perspektiv-review

For each perspective, identify concerns, risks, and recommendations, or briefly explain why it does not apply to the change.

Perspektiv 1: Arkitektur

  • Passer dette i Navs overordnede arkitektur?
  • Finnes det enklere alternativer?
  • Introduserer det unødvendig kompleksitet?
  • Er det i tråd med team-autonomi-prinsippet?
  • Gjenbruker det eksisterende plattform-kapabiliteter?
  • Navs prinsipper: Team First, essential complexity, Product Development

Perspektiv 2: Sikkerhet

Dypere sikkerhetsgjennomgang: For trusselmodellering, OWASP-sjekklister og compliance-vurderinger, delegér til @security-champion-agent som har komplett sikkerhetsmateriale.

  • Hvilke data behandles? Klassifiseringsnivå?
  • Er autentisering og autorisasjon riktig?
  • Er tilgangsstyring minimalt nødvendig (least privilege)?
  • Er PII beskyttet (logging, lagring, transit)?
  • Følger dette Navs Golden Path for sikkerhet?

Perspektiv 3: Plattform

  • Fungerer dette på Nais (Kubernetes/GCP)?
  • Er ressurskrav realistiske?
  • Er observerbarhet ivaretatt?
  • Er CI/CD-pipeline enkel og vedlikeholdbar?
  • Er det avhengigheter til on-prem eller legacy?

Perspektiv 4: Migrasjon (kun ved endring av eksisterende system)

  • Er endringen bakoverkompatibel? Kan gammel kode kjøre med nytt skjema?
  • Finnes det en rollback-plan som ikke medfører datatap?
  • Er det definert exit criteria for når migreringen er ferdig?
  • Er feature toggle satt opp for gradvis utrulling?
  • Er berørte konsumenter (andre team/tjenester) identifisert og informert?
  • Er det satt opp rekonsiliering for å verifisere datakonsistens?
  • Er dekommisjoneringsplan for gammel løsning definert?
  • Er migrasjons-observerbarhet på plass (gammel vs ny path, avviksteller)?

Perspective 5: User experience

  • Who is affected, and what task must they be able to enter, complete, and check?
  • How does each alternative affect that workflow, including navigation, necessary information, feedback, recovery, and accessibility?
  • For a plan, inspect the proposed flow and assumptions. For implemented UI, exercise the task on the rendered interface; record what was inspected and any missing evidence.
  • Judge reading, comparison, and monitoring views by those jobs. Correct components and passing tests alone do not establish usability.

Steg 3: Alternativer

Identifiser minst to alternativer:

AlternativFordelerUlemperNav-vurdering
A: [valgt].........
B: [forkastet].........
C: Gjøre ingenting.........

Steg 4: Generer ADR

Bruk malen fra adr-template.md.

markdown
# ADR-{nummer}: {tittel}

**Dato:** {dato}
**Status:** Foreslått | Godkjent | Forkastet | Erstattet av ADR-{n}
**Beslutningstakere:** {team/personer}

## Kontekst

{Hva er problemet/muligheten? Hvorfor må vi ta en beslutning nå?}

## Beslutning

{Hva har vi besluttet å gjøre?}

## Alternativer vurdert

### Alternativ A: {navn} (valgt)
- **Fordeler:** ...
- **Ulemper:** ...
- **Nav-vurdering:** ...

### Alternativ B: {navn}
- **Fordeler:** ...
- **Ulemper:** ...
- **Nav-vurdering:** ...

## Nav-spesifikke vurderinger

### Sikkerhet
- Dataklassifisering: {nivå}
- Auth-mekanisme: {valgt mekanisme}
- PII-håndtering: {strategi}

### Plattform
- Nais-konfigurasjon: {endringer}
- Ressursbehov: {estimat}
- Observerbarhet: {strategi}

### Team-påvirkning
- Berørte team: {liste}
- Migrasjonsstrategi: {plan}
- Tilbakerulle-strategi: {plan}

### Brukeropplevelse
- Berørte brukere og oppgave: {arbeidsflyt, eller hvorfor det ikke gjelder}
- Effekt av beslutningen: {informasjon, samhandling, universell utforming og feilretting}
- Grunnlag og hull: {foreslått flyt eller utprøvd grensesnitt, åpne antakelser}

### Migrasjon (ved endring av eksisterende system)
- Bakoverkompatibilitet: {vurdering}
- Utrullingsstrategi: {big bang / gradvis / parallell}
- Feature toggle: {toggle-navn og strategi}
- Rollback-trigger: {hva utløser rollback}
- Exit criteria: {når er migreringen ferdig}
- Dekommisjonering: {plan for gammel løsning}

## Konsekvenser

### Positive
- ...

### Negative
- ...

### Risiko
- ...

## Aksjonspunkter

- [ ] {konkret oppgave med eier}

Steg 5: Aksjonspunkter

Avslutt alltid med konkrete aksjonspunkter:

  • Implementer beslutningen
  • Oppdater Nais-manifest
  • Oppdater CI/CD-pipeline
  • Informer berørte team
  • Oppdater dokumentasjon
  • Sett opp observerbarhet

Navs arkitekturprinsipper

Se nav-principles.md for detaljerte prinsipper.

Kort oppsummert:

  1. Team First — Autonome team med sirkel av autonomi
  2. Essensiell kompleksitet — Unngå accidental complexity
  3. Produktutvikling — Kontinuerlig utvikling, produktorganisert gjenbruk
  4. DORA-metrikker — Mål og forbedre team-ytelse
  5. Architecture Advice Process — Søk råd fra berørte parter, men ta beslutningen selv

Teknisk gjeld-vurdering

Bruk denne strukturen til å identifisere, dokumentere og prioritere teknisk gjeld. Spesielt nyttig ved modernisering og refaktorering.

Identifisering

Still disse spørsmålene for å finne teknisk gjeld:

OmrådeSpørsmål
AvhengigheterFinnes det utdaterte avhengigheter med kjente sårbarheter?
ArkitekturEr det moduler med uklare ansvarsområder eller tett kobling?
KodeFinnes det duplisert logikk, store klasser/filer, eller manglende abstraksjoner?
TesterEr testdekningen lav i kritisk forretningslogikk? Mangler integrasjonstester?
InfrastrukturBrukes utdatert base image, mangler observerbarhet, eller er CI/CD treg?
DokumentasjonMangler ADR-er for viktige beslutninger? Er runbooks utdaterte?
SikkerhetEr auth-mekanismen oppdatert? Logges PII?

Prioriteringsmodell

Prioriter gjeldsposter med tre dimensjoner:

Prioritet = Alvorlighetsgrad × Frekvens × Nedslagsfelt

Alvorlighetsgrad (1-3):
  1 = Irriterende men håndterbart
  2 = Bremser utvikling merkbart
  3 = Blokkerer ny funksjonalitet eller utgjør sikkerhetsrisiko

Frekvens (1-3):
  1 = Sjeldent (årlig)
  2 = Regelmessig (månedlig)
  3 = Ofte (ukentlig eller daglig)

Nedslagsfelt (1-3):
  1 = Én komponent
  2 = Flere komponenter i samme tjeneste
  3 = På tvers av tjenester/team

Output: Gjeldstabell

markdown
| # | Gjeldspost | Alv. | Frekv. | Nedslagsfelt | Prioritet | Anbefalt tiltak |
|---|-----------|------|--------|-------------|-----------|----------------|
| G1 | Utdatert auth-bibliotek | 3 | 3 | 2 | 18 | Oppgrader til nyeste navikt/oasis |
| G2 | Manglende testdekning i beregningsmodul | 2 | 3 | 1 | 6 | Skriv karakteriseringstester |
| G3 | Duplisert valideringslogikk | 1 | 2 | 2 | 4 | Ekstraher til felles modul |

Leverabel: Gjeldstabell med prioritert rekkefølge og anbefalt tiltak per post.

Boundaries

✅ Always

  • Inkluder minst to alternativer (inkludert «gjøre ingenting»)
  • Review all five perspectives; explain any that do not apply.
  • Dokumenter Nav-spesifikke vurderinger (auth, data, plattform)
  • Avslutt med konkrete aksjonspunkter

⚠️ Ask First

  • ADR som påvirker andre teams tjenester
  • Beslutninger som avviker fra Navs standardmønstre
  • Introduksjon av ny teknologi

🚫 Never

  • Fatt beslutning uten å vurdere sikkerhet
  • Ignorer plattform-konsekvenser
  • Hopp over alternativer — det finnes alltid minst to valg

Review contract

Axes: the five perspectives in Steg 2 — architecture, security, platform, migration, and user experience. State why any axis does not apply. The verdict is on the chosen alternative, and it gates the ADR Status field: BLOCK means the ADR is not approvable as written.

The review ends here, in this shape. A review with no output section has not run.

Judge primary evidence. The diff, the file, the EXPLAIN output, the rendered page — never your own summary of the change, and never your memory of what you meant to write. Resolve the base first, then cover committed, staged, unstaged and untracked changes. Read untracked files in full: diff output omits them.

Every axis reports. Each applicable axis produces at least one finding, or one line saying what it inspected and what that evidence does not prove. For an inapplicable axis, state why. An axis that says nothing has not looked.

Report what you inspected. "No findings in the two files I opened" and "no findings in the change" are different claims, and only the first one is ever true.

Inspected:     <files, queries, URLs, viewports actually opened>
Not inspected: <in scope, not examined, and why>
Findings:      <n blocking, n concerns>
Verdict:       BLOCK | CONCERNS | CLEAN
  • BLOCK — at least one finding that, shipped as written, risks data loss, a security or privacy breach, a production incident, a wrong answer to a user, or prevents an intended user from completing a critical task.
  • CONCERNS — no blocking finding, but at least one a maintainer should fix or answer first.
  • CLEAN — every axis inspected against primary evidence, nothing at either bar. CLEAN claims only the axes above and the files on the Inspected line, and it is wrong if a defect is later found in them.

Not a review: LGTM; restating what the change does; cosmetic findings only; reading the changed lines without the code they call.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Nav Architecture Review AI skill do?

Generer Architecture Decision Records (ADR) med flerperspektiv-review tilpasset Nav

Why use Nav Architecture Review on TypingMind?

Because you install it once and use it with any model. Nav Architecture Review is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Nav Architecture Review in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/navikt/copilot/tree/main/skills/nav-architecture-review. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Nav Architecture Review?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Nav Architecture Review?

As many as you like. As long as a model supports skills, you can use Nav Architecture Review with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Nav Architecture Review AI skill free?

Yes. It is published on GitHub by navikt under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇