Contributing to Dendrite

Everyone is welcome to contribute to Dendrite! We aim to make it as easy as possible to get started.

Contribution types

We are a small team maintaining a large project. As a result, we cannot merge every feature, even if it is bug-free and useful, because we then commit to maintaining it indefinitely. We will always accept:

  • bug fixes
  • security fixes (please responsibly disclose via before creating pull requests)

We will accept the following with caveats:

  • documentation fixes, provided they do not add additional instructions which can end up going out-of-date, e.g example configs, shell commands.
  • performance fixes, provided they do not add significantly more maintenance burden.
  • additional functionality on existing features, provided the functionality is small and maintainable.
  • additional functionality that, in its absence, would impact the ecosystem e.g spam and abuse mitigations
  • test-only changes, provided they help improve coverage or test tricky code.

The following items are at risk of not being accepted:

  • Configuration or CLI changes, particularly ones which increase the overall configuration surface.

The following items are unlikely to be accepted into a main Dendrite release for now:

  • New MSC implementations.
  • New features which are not in the specification.

Sign off

We require that everyone who contributes to the project signs off their contributions in accordance with the Developer Certificate of Origin. In effect, this means adding a statement to your pull requests or commit messages along the lines of:

Signed-off-by: Full Name <email address>

Unfortunately we can’t accept contributions without a sign-off.

Please note that we can only accept contributions under a legally identifiable name, such as your name as it appears on government-issued documentation or common-law names (claimed by legitimate usage or repute). We cannot accept sign-offs from a pseudonym or alias and cannot accept anonymous contributions.

If you would prefer to sign off privately instead (so as to not reveal your full name on a public pull request), you can do so by emailing a sign-off declaration and a link to your pull request directly to the Foundation at Once a private sign-off has been made, you will not be required to do so for future contributions.

Getting up and running

See the Installation section for information on how to build an instance of Dendrite. You will likely need this in order to test your changes.

Code style

On the whole, the format as prescribed by gofmt, goimports etc. is exactly what we use and expect. Please make sure that you run one of these formatters before submitting your contribution.


Please make sure that the comments adequately explain why your code does what it does. If there are statements that are not obvious, please comment what they do.

We also have some special tags which we use for searchability. These are:

  • // TODO: for places where a future review, rewrite or refactor is likely required;
  • // FIXME: for places where we know there is an outstanding bug that needs a fix;
  • // NOTSPEC: for places where the behaviour specifically does not match what the Matrix Specification prescribes, along with a description of why that is the case.


We use golangci-lint to lint Dendrite which can be executed via:

golangci-lint run

If you are receiving linter warnings that you are certain are spurious and want to silence them, you can annotate the relevant lines or methods with a // nolint: comment. Please avoid doing this if you can.

Unit tests

We also have unit tests which we run via:

DENDRITE_TEST_SKIP_NODB=1 go test --race ./...

This only runs SQLite database tests. If you wish to execute Postgres tests as well, you’ll either need to have Postgres installed locally (createdb will be used) or have a remote/containerized Postgres instance available.

To configure the connection to a remote Postgres, you can use the following enviroment variables:

POSTGRES_DB=postgres # the superuser database to use

In general, we like submissions that come with tests. Anything that proves that the code is functioning as intended is great, and to ensure that we will find out quickly in the future if any regressions happen.

We use the standard Go testing package for this, alongside some helper functions in our own test package.

Continuous integration

When a Pull Request is submitted, continuous integration jobs are run automatically by GitHub actions to ensure that the code builds and works in a number of configurations, such as different Go versions, using full HTTP APIs and both database engines. CI will automatically run the unit tests (as above) as well as both of our integration test suites (Complement and SyTest).

You can see the progress of any CI jobs at the bottom of the Pull Request page, or by looking at the Actions tab of the Dendrite repository.

We generally won’t accept a submission unless all of the CI jobs are passing. We do understand though that sometimes the tests get things wrong — if that’s the case, please also raise a pull request to fix the relevant tests!

Running CI tests locally

To save waiting for CI to finish after every commit, it is ideal to run the checks locally before pushing, fixing errors first. This also saves other people time as only so many PRs can be tested at a given time.

To execute what CI tests, first run ./build/scripts/; this script will build the code, lint it, and run go test ./... with race condition checking enabled. If something needs to be changed, fix it and then run the script again until it no longer complains. Be warned that the linting can take a significant amount of CPU and RAM.

Once the code builds, run Sytest according to the guide in docs/development/ so you can see whether something is being broken and whether there are newly passing tests.

If these two steps report no problems, the code should be able to pass the CI tests.

Picking things to do

If you’re new then feel free to pick up an issue labelled good first issue. These should be well-contained, small pieces of work that can be picked up to help you get familiar with the code base.

Once you’re comfortable with hacking on Dendrite there are issues labelled as help wanted, these are often slightly larger or more complicated pieces of work but are hopefully nonetheless fairly well-contained.

We ask people who are familiar with Dendrite to leave the good first issue issues so that there is always a way for new people to come and get involved.

Getting help

For questions related to developing on Dendrite we have a dedicated room on Matrix where we’re happy to help.

For more general questions please use