Cloud API Platform Docs logoCloud API
Cloud APIdocs

Documentation · API v2

Cloud API Platform Docs

Everything you need to authenticate, call endpoints, and ship.

A structured reference for integrating with the platform. Read the overview, set up credentials, and make your first request in minutes.

Start building~5 min to first request

01 / docs

Overview

Everything you need to go from first request to production integration.

The platform exposes a versioned REST API over HTTPS with JSON payloads. These docs are ordered the way you build: understand the model, authenticate, call endpoints, then resolve edge cases. Each section is self-contained and linkable, so you can jump straight to what you need.

  • Clear structure

    Logical order, stable anchors, no dead ends.

  • Secure auth model

    Scoped API keys and bearer tokens.

  • Predictable endpoints

    Consistent resources, errors, and pagination.

  • Implementation-ready

    Copy-paste examples for every call.

docs/structure4 areas
/overview
Platform concepts and how the docs are organized.
/authentication
Create keys, send tokens, and manage scopes.
/endpoints
Resource reference with parameters and responses.
/faq
Answers to common integration questions.

02 · Security

Authentication

Every request is authorized with a bearer token scoped to a project and environment.

  1. 1

    Obtain credentials

    Create an API key in the console, then exchange it for a short-lived access token.

  2. 2

    Send the token

    Include it in the Authorization header over HTTPS on every call.

  3. 3

    Validate responses

    Handle 401 by refreshing the token and 403 by checking scopes.

Example request

curl https://api.example.cloud/v1/projects \
  -H "Authorization: Bearer $API_TOKEN" \
  -H "X-Environment: sandbox"

# 200 OK   → token valid
# 401      → refresh token
# 403      → missing scope

Best practices

  • Store securely. Keep tokens in a secrets manager, never in client code.
  • Rotate regularly. Rotate keys every 90 days and revoke unused ones.
  • Separate environments. Use distinct keys for sandbox and production.

03 / Reference

Endpoints

The API is organized into four resource groups. All endpoints are served from https://api.example.cloud/v1.

Accounts

/accounts

Organization profile, members, and role assignments.

Operations

  • GET/accounts/{id}
  • PATCH/accounts/{id}
  • GET/accounts/{id}/members

Projects

/projects

Create and manage isolated project environments and keys.

Operations

  • GET/projects
  • POST/projects
  • DELETE/projects/{id}

Usage

/usage

Metered request counts, quotas, and billing periods.

Operations

  • GET/usage
  • GET/usage/quotas
  • GET/usage/export

Webhooks

/webhooks

Subscribe to events; signed payloads with retry delivery.

Operations

  • POST/webhooks
  • GET/webhooks/{id}/deliveries
  • DELETE/webhooks/{id}

Consistency: every endpoint accepts and returns JSON, uses cursor-based pagination, and returns errors in a shared { error: { code, message } } shape.

05 / Support

Frequently asked questions

Short answers to the questions teams ask most before shipping their first integration.

What are the rate limits?
Standard keys allow 600 requests per minute per project. Every response includes X-RateLimit-Remaining and X-RateLimit-Reset headers. Exceeding the limit returns 429.
How do sandbox and production environments differ?
Sandbox uses sandbox.api with test keys and mirrors production behavior without billing. Keys are scoped per environment and cannot be used across them.
How should I handle retries?
Retry 429 and 5xx responses with exponential backoff and jitter. Send an Idempotency-Key header on write requests so retries never duplicate work.
How is the API versioned?
Versions are date-based and pinned per request via the API-Version header. Breaking changes ship only in new versions, and deprecated versions remain supported for at least 12 months.
What support response times can I expect?
Community plans receive replies within two business days. Paid plans get responses within 4 hours, and critical production incidents on enterprise plans are acknowledged within 30 minutes, 24/7.

Still have questions? The quickest answer is often a first request.

Start building