Trying it locally

You do not need your company’s Confluence to try Confluence Explorer or to work on it. The source repository can start a real Confluence Data Center on your own machine, fill it with a small set of test spaces, and run the app against it. Nothing you do there reaches any other Confluence, and your own settings in ~/.confluence-explorer are left alone.

What you need

• Node.js 22 or later, and Corepack, which comes with Node.js and installs the right version of pnpm for you.
• podman or docker, to run Confluence and its database in containers. Confluence may use up to 4 GB of memory, and its database a little more.
• git, to get the source.

Get the source

corepack enable
git clone https://github.com/SimplySF/simply-confluence-explorer.git
cd simply-confluence-explorer
corepack install
pnpm install

corepack enable is needed once per machine. To try a change that is still in a pull request, switch to its branch first, for example git switch feat/my-change, then run pnpm install again.

Start a Confluence of your own

pnpm run testenv:start

The first start downloads Confluence and Postgres and sets them up, which takes a few minutes; later starts take about a minute. It finishes by creating four test spaces, PLAT, OPS, PEOPLE and ENG, with pages, labels, attachments and .confluenceignore rules, so every feature has something to work on. The administrator account is admin with the password admin, and Confluence itself is at http://localhost:8090 if you want to look at the pages there.

Confluence runs on Atlassian’s developer licence, which lasts three hours. pnpm run testenv:status shows when it runs out; after that every write is refused, and pnpm run testenv:down followed by pnpm run testenv:start restarts it with a new licence.

Run the app against it

export $(node test-env/testenv.mjs env | xargs)
pnpm run dev

The first line points the app at your local Confluence, with a token it created, and gives it a data folder of its own in test-env/.state/data. (node test-env/testenv.mjs env prints the four variables it sets, if your shell needs them set another way.) Open http://localhost:5173. The app is already connected, so you can sync a space, read pages, review them and push changes straight away. Changes to the browser side of the code appear as you save them; changes to the server need pnpm run dev to be started again.

For the AI features, choose a provider under Settings → AI provider. The settings are saved in the test data folder, not in your own.

To try the app exactly as people install it, build it and run the packed copy instead, keeping the same variables:

pnpm run build
cd packages/simply-confluence-explorer
pnpm pack
npx ./simplysf-simply-confluence-explorer-<version>.tgz

and open the address it prints.

Run the tests

pnpm run build
pnpm test
pnpm run test:e2e

pnpm test is the fast suite, which needs no Confluence. pnpm run test:e2e runs against your local Confluence, and skips itself when none is running.

Stop or start again

command	what it does
pnpm run testenv:status	Shows whether Confluence is running and when its licence runs out.
pnpm run testenv:down	Stops Confluence and keeps its pages for next time.
pnpm run testenv:reset	Deletes everything, so the next start begins with fresh pages.

Pages you change while trying the app stay changed until you reset. The test environment’s own README, test-env/README.md, explains how it works.

How this is checked

The e2e workflow (.github/workflows/e2e.yml) runs pnpm run testenv:start and the end-to-end suite on every pull request, so the steps above are run by CI as well as described here.