Build on InfraFlow

The API speaks JSON over HTTPS at https://api.infraflow.app. Every path starts with /v1. Almost everything lives under a workspace, named by its slug: /v1/orgs/{org}/….

curl https://api.infraflow.app/v1/orgs \
  -H "Authorization: Bearer $INFRAFLOW_TOKEN"

• Requests carry a JSON body with Content-Type: application/json. Unknown fields are ignored.
• Responses are JSON. A call that has nothing to return answers 204 with no body.
• Times are RFC 3339 in UTC: 2026-01-15T09:30:00Z. Identifiers are UUIDs; things you name (workspaces, projects, services, buckets) are addressed by that name.
• Errors are problem documents with a stable code.

The machine-readable description is at https://api.infraflow.app/v1/openapi.json (OpenAPI 3.1): generate a client from it in any language.

Each product gives what you create an address under its own zone. A resource named shop is reached at:

GET /v1/platform returns these zones, so nothing has to be hard-coded.

• Platform2 endpointsFacts about this deployment that clients need before anything else: the hostname zones, the storage endpoint, which sign-in methods exist, and whether the platform is open.
• Workspaces10 endpointsA workspace (an organisation, in the API's words) owns everything else. An API token belongs to exactly one.
• Members4 endpointsWho belongs to a workspace, and with which role: owner, admin, member or viewer.
• API tokens3 endpointsCredentials for scripts and CI. A token acts within its workspace, with the role it was given.
• Projects and environments7 endpointsA project groups services. Each environment of a project has its own private network, in which services reach one another by name.