Quickstart

revcat auth login

revcat opens your browser for the OAuth flow, you authorize against RevenueCat, and the tokens land in ~/.revcat/config.json (mode 0600).

2. Bind a project to your repo

cd ~/your-repo
revcat init

revcat init

revcat init lists projects you can access, prompts for one (and optionally apps), then writes:

revcat.toml (committed): records project_id + apps so a git clone documents which RC project this repo belongs to.
.revcat/config.json (gitignored, mode 0600): copies your credential into the directory so subsequent commands and any agent or sandbox in the directory inherit it without touching ~/.revcat/config.json.

.revcat/ is auto-appended to .gitignore.

3. Verify

revcat auth doctor
revcat doctor

Both should report OK for every check.

4. Run your first command

Inside the repo, project context is automatic:

revcat metrics overview
revcat subscribers info app_user_123
revcat publish offering pro --paywall ./paywalls/pro.json

Multiple accounts

Juggling several RC accounts? Log in to each with a different --name and switch:

revcat auth login --name work
revcat auth login --name personal
revcat auth use personal                 # default for global commands
revcat --profile work auth status        # one-shot override

When you revcat init inside a repo, it copies whichever profile is active at that moment. If you need to switch credentials for an existing repo, rerun revcat init --force.

Headless / CI

For fresh sandboxes with no browser:

export REVCAT_REFRESH_TOKEN=rtk_...
export REVCAT_PROJECT_ID=proj_...
revcat offerings list

revcat synthesizes a virtual profile, refreshes tokens in-memory, and skips the login flow entirely. Pull the refresh token from your CI secret manager.