GitHub Sync

A scheduled fetcher that pulls Playwright report artifacts out of a GitHub Actions workflow and uploads them as native reports. Use it when your CI already produces playwright-report artifacts and you'd rather not edit the workflow file again because of security policies or religious beliefs.

Configured from Settings -> Schedules -> GitHub Sync. You can run multiple syncs side by side, each pointing at a different repo with its own schedule and token. They don't interfere with each other.

Setting up a sync

Each entry has these fields:

Field	What you put there
Name	Whatever helps you recognize it in the list.
Repository	`owner/name`.
Workflow file	Just filename, e.g. `playwright.yml`.
GitHub token	Optional per-config token. Stored encrypted. If blank, falls back to the `GITHUB_TOKEN` env var.
Start date & time	Workflow runs before this point get ignored. Helpful when you do not want to sync everything.
Cron schedule	Standard cron expression. Each config has its own job.
Artifact name regex	Filters which artifacts to upload. Capture groups become `${match1}`, `${match2}`, etc. below.
Project name template	Produces the report's `project` field. Mix literal text with placeholders.
Report title template	Produces the report's display title.
Enabled	Unchecking does not schedule the job.

Template placeholders

Available everywhere the UI says "template":

Placeholder	What it means
`${match1..N}`	Capture groups from the artifact name regex
`${branch}`	`head_branch` from the workflow run
`${runDate}`	First 10 chars of the run's `created_at` (YYYY-MM-DD)
`${runId}`	GitHub workflow run id
`${artifactName}`	Raw artifact name returned by GitHub
`${workflowName}`	Workflow display name
`${workflowFile}`	Workflow filename
`${repo}`	`owner/name` from the config

Example

You have same workflow for dev and qa environments.
Artifacts are playwright-report-dev-chrome and playwright-report-qa-chrome, so we can match with regex:
^playwright-report-(.+)-(.+)$. match1 would be your env name, match2 would be the browser. So we can use it for project template: ${match1}:${match2}, and have projects dev:chrome and qa:chrome.

What permissions the token needs

Read-only.

Fine-grained PAT (recommended): Actions: Read-only on the target repo. Metadata: Read-only is granted automatically.
Classic PAT: repo for private repos, or public_repo for public. Yes, that's more access than necessary; that's classic PAT for you.

If you leave the token field blank in the UI, the server falls back to the GITHUB_TOKEN env var. Convenient for a "shared" setup; less convenient if it turns out to be over-scoped, so prefer per-config PATs in anything serious.

Local dev with the gh CLI:

bash

export GITHUB_TOKEN="$(gh auth token)"

What you'll see

On each cron tick, the sync lists workflow runs since the start date, filters artifacts by the regex, skips anything it's already imported (so it's safe to re-run), downloads what's left, and uploads each as a native report. Templates resolve per artifact. Storage backend (fs / s3 / azure) is whatever you configured globally; sync is just a producer.

Common setups

Per-browser projects on one workflow. Use ^playwright-report-(chrome|firefox|webkit)$ and template ${match1}. Three projects emerge automatically, no per-browser config.

Branch-segregated reports. Template ${branch} (or ${match1}:${branch}) so PR runs and main runs land in different projects. PR noise stops drowning your main-branch trends.

Backfill. Set Start date far in the past and save. The next tick imports everything matching since then.

GitHub Sync ​

Setting up a sync ​

Template placeholders ​

Example ​

What permissions the token needs ​

What you'll see ​

Common setups ​

See also ​