Apache-2.0 · Self-hosted · No per-user fees

The Scrum Guide, enforced.

Scrumooth is the gatekeeper, not the note-taker. It turns the 2020 Scrum Guide's rules into gates the backend holds you to — a Sprint cannot close without its Review and Retrospective, and only Developers size the work. It is the Scrum Guide enforcement layer your tracker does not have.

No install, no signup. The demo runs on mock data in your browser and resets on refresh.

  • 68 backend-enforced gates
  • One Docker Compose stack
  • Audit log for every state change
scrumooth · gatekeeper
  • close sprint 7refusedReview and Retrospective not recorded
  • size backlog itemrefusedonly Developers may size the work
  • cancel sprint 7refusedProduct Owner only, while the Sprint is ACTIVE

Three of the 68 refusals the backend can return. Every one holds in the backend service layer, so neither the interface nor a direct API call gets around it.

Recording is not enforcement.

Recording is genuinely useful, and the tools you already own do it well. But a record is a description, not a decision. The 2020 Scrum Guide is full of rules a tool could hold you to — and unless a tool is built to enforce them, those rules stay advisory: a shared understanding the team is trusted to remember. So the question worth asking of any Scrum tool is not what it draws, but which rules it will refuse to break.

What recording gives you

Accurate — after the fact

  • Boards, charts and a complete history of what happened.
  • Process lives in configuration — a real strength, because the tool adapts to the team.
  • A Sprint closed before its Retrospective is logged accurately, afterwards.

A rule you can reconfigure is a setting, not a rule.

What enforcement gives you

Decisive — before it slips

  • The 2020 Scrum Guide embedded as executable code.
  • Enforced server-side, where neither the interface nor a direct API call can bypass it.
  • Configurations that break the Guide are never offered — so there is nothing to switch off.

The refusal is the product.

It is Friday. The Sprint is due to end, the Increment is deployed — and the Retrospective was never scheduled.

In a tool that only records, the Sprint closes and the Retrospective slips to next week — precisely the failure the Guide's final event exists to prevent. Scrumooth refuses the close until both the Review and the Retrospective are recorded. The team then runs the Retrospective — or stops and discusses why not, which is the version of that decision the Guide expects a team to make consciously.

An illustration of the difference. Fewer process debates. More time shipping working software.

If you write software, the closest analogy is a linter for your Scrum process — except a linter reports a violation, and a gate refuses it. None of this is a feature the tools you already use are missing: it is a trade-off they have already made the other way.

What Scrumooth enforces

These are gates, not warnings or hints. In every case the answer is no — and every answer holds in the backend service layer, so a frontend shortcut cannot get around it. The backend contract holds 68 refusals: 39 from the 2020 Scrum Guide, 5 complementary practices and 24 process-integrity gates — counted from that one contract and verified in CI, so the totals cannot drift from what the code enforces.

Scrum Guide rules Scrumooth refuses to break, and how the backend answers them
A 2020 Scrum Guide rule, asked as a questionScrumooth's answer
Can a Sprint be closed before its Review and Retrospective?RefusedSprint completion is refused until both events are recorded.
Can an item be called Done without its Definition of Done?RefusedCompleting a Sprint never marks items Done — each item must pass its Definition of Done checklist.
Can a team hold more than one Product Owner or Scrum Master?RefusedAdding a second holder of either role is refused.
Can a team grow past Scrum Team size?RefusedTeam size is capped by TEAM_MAX_SIZE, default 10.
Can someone other than a Developer size the work?RefusedOnly Developers can size Product Backlog items — every other role receives 403 Forbidden.
Can the Product Owner or Scrum Master author the Daily Scrum?RefusedOnly Developers can author or join the daily record; the Product Owner and Scrum Master observe.
Can a Sprint be cancelled by anyone but the Product Owner?RefusedCancellation is Product-Owner-only, and only while the Sprint is ACTIVE.
Can a delivered Increment be rewritten?RefusedDelivered Increments are locked against further edits.

The eight rows above are a guided tour, not the catalogue: every GATE_* refusal, with the HTTP status it is returned with, is the gate rejections reference — the single source of truth for what this product claims.

The other 29 are declared as what they are — 5 complementary-practice gates for the product's own Definition of Ready, labelled a practice in the interface rather than a Guide rule and left to the team's Scrum Master to shape or retire, plus 24 process-integrity gates that keep a team's artifacts readable only by its members, and the Scrum Master's own material only by the Scrum Master. Three classes, kept apart on purpose, so the Guide's authority is never borrowed for a rule the Guide does not contain.

Where Scrumooth deliberately does not enforce anything

The Guide asks for self-management in exactly these places, so Scrumooth does not decide for the team.

  • The Retrospective Prime DirectiveLeft to the facilitator, where it belongs.
  • Event timeboxesSurfaced through a shared team timer rather than forcibly terminating an event.

The catalogue is the entire claim. If a refusal is not in it, Scrumooth does not enforce it — and because a configuration that breaks the 2020 Scrum Guide is never offered, the refusal is the product.

Watch it refuse somethingRuns in your browser on mock data.

The whole Sprint, rule by rule

Every event, artifact and commitment of the 2020 Scrum Guide, with the rule it holds attached to it — and every card backed by a gate the backend enforces. This is the tour, not the catalogue; the complete list is the gate rejections reference.

Product Goal

Strategic alignment and goal tracking — the commitment the backlog serves.

EnforcedOnly the Product Owner creates or edits one, only one may be ACTIVE, and a Sprint cannot start until it is linked to one

Product Backlog

MoSCoW prioritisation (Must, Should, Could, Won't).

EnforcedOnly Developers size the work, and only the Product Owner orders it and sets its band

Sprint Planning

Configurable Sprint durations and capacity planning.

EnforcedOnly Developers save the Sprint Backlog, a Sprint cannot start until planning attendance is recorded with the Product Owner and a Developer, and the plan must fit the recorded capacity

Sprint Execution

Interactive Kanban board with drag-and-drop.

EnforcedOnly the Product Owner can cancel, and only while the Sprint is ACTIVE

Daily Scrum

Shared daily record, with impediment surfacing.

EnforcedOnly Developers author it — the Product Owner and Scrum Master observe

Impediment

Blocker identification and resolution tracking.

EnforcedA Sprint cannot close before its Impediments are resolved

Increment

Product increment management.

EnforcedThe moment an item meets the Definition of Done, an Increment is born

Sprint Review

Review management, stakeholder feedback, and backlog adjustment.

EnforcedA Sprint cannot close before its Review is recorded, and a Sprint with a Goal cannot complete without the team’s verdict on it

Sprint Retrospective

Team reflection and tracked improvement.

EnforcedA Sprint cannot close before its Retrospective is recorded, and an improvement already linked to a Product Backlog item cannot be unlinked

Also included

  • Role-based workflow engine with gated state transitions
  • Definition of Done, plus a Definition of Ready the interface labels a complementary practice rather than a Guide artifact
  • Dedicated, compliance-separated audit log
  • Dashboard and reporting
  • Team health check against the five Scrum values
  • Shared event timeboxes — surfaced, never force-closed
  • Data export and erasure rights, with consent tracking
  • Interface in English, German, Spanish, French and Italian

Who it's for — and who it isn't

Scrumooth is built for one situation in particular: engineering-led organisations that must be able to show how a Sprint was actually run — and it serves them by refusing, not recording, because the record is a by-product of the gate rather than a description written after it. Process data never leaves their own infrastructure, which is why regulated industries, their suppliers, and public-sector teams are its fit.

Scrumooth is for you if…

  • You are a Scrum Master or Product Owner whose team finds it hard to hold to the 2020 Scrum Guide, and you want the tool to refuse the drift instead of quietly allowing it.
  • You lead an engineering team that wants to self-host its process data for privacy, compliance or data-sovereignty reasons.
  • You need a defensible, auditable record of how each Sprint was actually run — who changed what, when, and under which role.
  • You want the Guide's boundaries encoded once, so new team members learn the process by using it.

Scrumooth is not for you if…

  • You want a general-purpose issue tracker, roadmap planner or Kanban board for non-Scrum work. Scrumooth refuses to be one.
  • You want every rule to be configurable. Scrumooth refuses configurations that break the 2020 Scrum Guide.
  • You want a fully managed SaaS. Scrumooth is self-hosted by design.
  • You need deep portfolio management, resource planning or financial tracking across many unrelated projects.
  • You follow a scaled framework that adapts the Guide for a wider organisation, or Scrum is not yet how your team works.

A tool that states what it will not do is easier to evaluate.

Why you can trust it

The people who approve software for regulated teams ask the same questions: where does the data live, who can reach it, and can we prove what happened. Scrumooth answers all three before it is deployed.

Your process data never leaves your infrastructure

Self-hosted by design, on your hardware, under your own policies and backups.

Data sovereignty ships with the product

GDPR data export, a 14-day deletion grace period and consent tracking are built in, not bolted on.

Every decision is auditable

Each role change and state transition is written to a dedicated, compliance-separated audit log.

Access is bounded

Concurrent sessions are capped, and the oldest sessions are revoked automatically.

For those who have to approve it internally

Deployment guidance, the security architecture and the vulnerability-reporting process are all documented in the repository — read them before you commit to anything.

Questions worth asking before you deploy

What exactly is Scrumooth?

A self-hosted, open-source web application for teams that run Scrum. It owns the Sprint lifecycle, the roles and the gates, and enforces the 2020 Scrum Guide in the backend service layer. It is deliberately not a replacement for your issue tracker: your tracker keeps your record, Scrumooth keeps your rules.

How is it different from Jira or Linear?

Different problem, different trade-off. Jira and Linear are general-purpose trackers, and flexibility is exactly what makes them good at that job. Scrumooth does not try to out-flex them: it enforces the 2020 Scrum Guide as written, and never offers a configuration that breaks it — so a Sprint cannot be closed before its Review and Retrospective are recorded. A question worth asking of any tracker: which rules will it refuse to break? If the honest answer is “that depends how you configured it”, that is a setting, not a rule. None of this is a capability those tools are missing: a gate you can switch off is a setting, not a rule, and flexibility is the trade-off that makes them good at what they do. Scrumooth makes the opposite one, deliberately.

Is Scrumooth free?

Yes. It is licensed under Apache-2.0: free to use, modify and distribute, with no per-user fees and no licence tiers. You pay only for the infrastructure you host it on.

Where does my data live?

On your own infrastructure. Scrumooth is self-hosted by design, so process data never leaves your servers. GDPR data export, a 14-day deletion grace period and consent tracking ship with the product, and every role change and state transition is written to a dedicated, compliance-separated audit log.

Can I turn a rule off?

No — and that is the product. A configuration that breaks the 2020 Scrum Guide is never offered, so there is no switch to find. If a refusal is not in the published catalogue of 68, Scrumooth does not claim it.

Do I have to migrate off my current tracker?

No. Scrumooth runs alongside the tools you already use and owns the part a tracker is not designed to hold: the Sprint lifecycle, the roles and the gates. Nothing about it conflicts with your existing record-keeping.

What if my team does not run Scrum exactly as the Guide describes?

Then it is probably not the right tool for you. Scrumooth enforces the 2020 Scrum Guide as written, for a single Scrum Team, and it is not intended for scaled frameworks that adapt the Guide for a wider organisation.

How quickly can I get it running?

One Docker Compose stack: reverse proxy, backend, frontend, PostgreSQL and scheduled backups. Clone the repository, copy the production environment file, run docker compose up -d, and you have a working instance on localhost.

Can I try it without installing anything?

Yes. The live demo runs in your browser with in-memory mock data — no backend, no database, no signup. Changes are local to your session and reset on refresh.

Which languages is the interface available in?

English, German, Spanish, French and Italian, with Scrum terminology sourced from the official Scrum Guide.

Jira and Linear are trademarks of their respective owners. They are named here only to identify the products being compared, and Scrumooth is not affiliated with, sponsored by or endorsed by either company. The comparison concerns the approach to process rules, not the quality or capability of those products.

Self-host it in one Compose stack

Running a second tool is a real cost — something else to deploy, secure, back up and keep fed. Scrumooth is deliberately the smallest system that can carry it: one stack, one database to look after.

  1. Clone the repositoryEverything ships in one monorepo: backend, frontend, shared types and docs.
  2. Copy the production env fileSet your database URL, a JWT secret, and the origin your team will reach.
  3. Bring the stack upReverse proxy, backend, frontend, PostgreSQL and scheduled backups start together.

Then open http://localhost. HTTPS is enabled by default on port 443. Full manual setup, prerequisites and operations guidance live in the deployment guide.

bash
git clone https://github.com/orbivort/scrumooth.git
cd scrumooth
cp packages/backend/.env.production.example packages/backend/.env.production
docker compose up -d

Make the rules hold.

Spin Scrumooth up on your own infrastructure, or try the demo first. Either way, the first thing you notice is what it refuses.

Judge a Scrum tool by the rules it keeps, not by the boards it draws.