Evidence, not assurance

Architecture governance that derives its own model.

Your cloud estate becomes the architecture model — collected, not drawn. Boards decide through an explicit, audited workflow. Conformance between what was approved and what is deployed runs continuously. Everything lands in a signed, chained evidence pack a third party can verify.

View on GitHub See how it works

docker compose up --build # one container, one SQLite file — usable in minutes

The Archigov overview: evidence and verification gauges, reviews by stage, open findings by severity and application, and the governance pipeline over a demo NHS estate.

References the frameworks you answer to

container — SQLite inside, nothing else to run

typed review flows, routed to the right board

AI features — each one optional, none of them decides

£0

per seat, ever — unlimited users on every option

From Your Estate to Signed Evidence

What happens between an Azure subscription and a decision a regulator can verify — collected, checked and decided on the record, with nothing drawn by hand.

01 · CollectResource Graph, Defender, Azure Policy and the backlog feed in — read-only.
02 · DeriveResources group into applications with provenance; evidence attaches where it belongs.
03 · CheckAdopted standards and patterns evaluate continuously; a contradiction becomes a finding.
04 · DecideThe board reviews: owners record positions per area, the chair declares the outcome.
05 · ProveEvery act lands in a SHA-256 audit chain, sealed by an Ed25519-signed evidence pack.

How It Works

Four ideas, enforced end to end. The product never asserts compliance — it shows evidence and who stands behind it.

01 · Derive

Derived, Not Drawn

Azure Resource Graph, the delivery backlog and your existing repositories populate the as-is model with provenance. Deterministic ArchiMate views come free.

02 · Check

Machine-Checkable Governance

Standards, principles, patterns and policies carry control references and, where possible, a rule evaluated continuously. Resource checks arrive as imported evidence, never rebuilt.

03 · Decide

Boards Decide, on the Record

Typed reviews route to boards with areas of responsibility. Decision makers record their own positions; the chair declares; minutes derive from the records.

04 · Prove

Signed Evidence, Continuously

Findings persist across runs, exceptions and conditions are tracked to resolution, and the evidence pack is Ed25519-signed against a SHA-256 audit chain.

AI Proposes. Humans Decide.

Eight AI features, each one switched on separately by your admin. Every feature keeps a deterministic fallback, every output is marked AI-generated, and nothing an AI writes approves, blocks or becomes the record without a named human's audited act.

Decision review — a critical test of the argument before it reaches the board
Pre-review — cited findings on a design; citations must resolve or they are discarded
Rule drafting — natural language to a checkable rule, validated and dry-run first
Evidence suggestions — missing evidence becomes a tracked assessment in one click

AI ReviewReview Argument

Draft: “Expose the Referral API directly to the internet with IP allow-listing — effectively as secure as a gateway.”

Argument needs work

ContradictionPAT-001 requires internet ingress through a governed gateway.

Unsupported claim“As secure as a gateway” — no evidence offered.

Risk not statedNo accepted risk for exposing an unhardened API.

Suggested evidenceSecurity assessment of the externally published surface.

Apply Tightened Argument

Open Source, Without the Rug-Pull

The core is AGPL-3.0. Run it yourself — even modified — for your own organisation with no obligations. Self-host from one container with SQLite inside; continuous Litestream replication if you want it. A managed UK/EU instance is available for teams that would rather not run it — hello@archigov.com.

Get the code

The commitment, made on day one: this project will never move to a licence more restrictive than AGPL-3.0. No BSL, no SSPL. Connectors, templates and the policy SDK will be published under Apache-2.0 as they split out, so integrating with Archigov never requires touching AGPL code.

Pricing

Flat per instance. Unlimited users — governance only works when everyone can take part, so there is no per-seat pricing, ever.

Self-Host

£0

AGPL-3.0, forever

The whole product — nothing held back
Unlimited users and boards
One container, your infrastructure
Community support on GitHub

Bring your own keys — integrations for OpenAI, Claude, Mistral, Ollama and any OpenAI-compatible endpoint
You pay your provider directly; no AI fees from us

Start on GitHub

Supported Self-Host

from £3,000

per instance, per year

Everything in Self-Host
Vendor support with response times
Upgrade and migration assistance
Scales with estate size

Bring your own keys — integrations for OpenAI, Claude, Mistral, Ollama and any OpenAI-compatible endpoint
You pay your provider directly; no AI fees from us

Talk to us

Managed UK/EU Instance

from £20,000

per instance, per year

We run it; you govern
Dedicated instance, continuously replicated
UK/EU data residency
Your Entra ID, your evidence keys

Included: EU-hosted models, no keys to manage
AI Premium add-on: frontier models under zero-retention contracts, from £3,600/yr
Additional usage from £0.20 per action
Your own model tenant (e.g. Azure OpenAI): £2,000/yr connector

Talk to us

Enterprise Self-Hosted

from £15,000

per instance, per year

Commercial licence where AGPL is barred
Everything in Supported Self-Host
Procurement-friendly paperwork

Bring your own keys or model tenant (e.g. Azure OpenAI) — integrations built in
You pay your provider directly; no AI fees from us

Talk to us

Compare Options

Every option includes every feature; the differences are who runs it, who supports it, and where AI comes from.
Detail	Self-Host	Supported Self-Host	Managed UK/EU Instance	Enterprise Self-Hosted
Product and licence
Price	£0	from £3,000 / year	from £20,000 / year	from £15,000 / year
Licence	AGPL-3.0	AGPL-3.0	AGPL-3.0 (we operate it)	Commercial licence
Features	All of them	All of them	All of them	All of them
Users and boards	Unlimited	Unlimited	Unlimited	Unlimited
Hosting and operations
Who runs it	You	You	We do	You
Where it runs	Your infrastructure	Your infrastructure	Dedicated UK/EU instance	Your infrastructure
Backups and replication	You (Litestream example included)	You, with our guidance	Continuous replication, run by us	You, with our guidance
Upgrades	You	Assisted upgrades and migrations	Done by us	Assisted upgrades and migrations
Sign-in	Local or your Entra ID	Local or your Entra ID	Your Entra ID	Your Entra ID
Evidence-signing keys	Generated on your instance	Generated on your instance	Per instance; public key published to you	Generated on your instance
Support
Support channel	Community (GitHub)	Vendor support	Vendor support	Vendor support
Response times	Best effort	Agreed response times	Agreed response times	Agreed response times
Procurement paperwork	—	Standard	Standard, incl. data processing terms	Procurement-friendly, incl. commercial licence
AI
AI provided by us	No	No	Yes — included	No
Provider and keys	Yours: OpenAI, Claude, Mistral, Ollama or any OpenAI-compatible endpoint	Yours: same integrations	Ours: EU-hosted models, no keys to manage	Yours: keys or your own model tenant (e.g. Azure OpenAI)
AI fees from us	None — you pay your provider	None — you pay your provider	Included allowance; additional usage from £0.20 per action	None — you pay your provider
Frontier models	With your own keys	With your own keys	AI Premium add-on, from £3,600 / year	With your own keys
Your own model tenant	Yes (your keys)	Yes (your keys)	£2,000 / year connector	Yes (your keys)

Prices are per instance and indicative; estates differ — write to hello@archigov.com and we will quote plainly.

Govern what you actually run — not what someone drew.

Start with the demo estate in five minutes, or talk to us about a managed UK/EU instance.

Get Started on GitHub hello@archigov.com