Your restoration data, over a REST API

Pull claims and payments into your accounting system, watch your playbooks from your own dashboards, or file a claim from the software you already run. JSON over HTTPS, an OpenAPI 3.1 document, and a hosted reference you can try calls from.

Included with every paid plan. Keys are minted by an organization admin in about a minute.

Quickstart

Base URL https://api.rubyrestore.ai/public/v1. Send your key as a bearer token. Every response is { data, meta, errors }; lists carry meta.next_cursor until the last page.

# 1. Who am I, and what can this key see?
curl https://api.rubyrestore.ai/public/v1/me \
  -H "Authorization: Bearer rr_live_..."

# 2. The latest runs of one playbook, newest first
curl "https://api.rubyrestore.ai/public/v1/playbooks/{id}/runs?limit=50" \
  -H "Authorization: Bearer rr_live_..."

What you can reach

Four resource groups. The full field list for each lives in the reference and is generated from the running API, so it cannot drift from what you get back.

Account

What the key can see: your organization, its workspaces, the integrations connected to each, and plan usage.

GET /me

GET /workspaces

GET /integrations

GET /usage

Playbooks

Your published playbooks and their live runs, so the automations Ruby runs for you are visible to your own systems.

GET /playbooks

GET /playbooks/{id}

GET /playbooks/{id}/runs

Claims and payments

For companies with a connected billing account: every claim with its committed amount, the settled and closed dates your accounting keys off, and the payments received on each.

Endpoints are listed in the API reference.

Claim submission

File a new claim from your own system. Documents are pulled from URLs you give; whether it files unattended or waits for a person is your connection's own setting.

Endpoints are listed in the API reference.

How access works

Included with every plan

API access comes with every paid Ruby plan, Starter and up. Calls never spend credits; each key has a generous per-minute and per-day rate limit that the response headers report.

Keys belong to your organization

An organization admin mints a key under Settings > API. It sees every workspace in the organization unless it is pinned to one, and it can be revoked at any moment.

Reads by default, writes opt-in

Every key can read. Filing a claim needs a write scope chosen when the key is created, so an accounting sync can never submit a claim by accident.

Built for a nightly sync

Lists are ordered by updated_at and paginate with a cursor. Store the last updated_at you saw, pass it back as updated_since, and upsert on id. No webhooks yet.

Looking for an AI client rather than code? The same data is available to ChatGPT, Claude and Cursor through RubyConnect.

Read the reference, mint a key, make a call.

Included with every plan.