OpenAPI spec

The machine-readable specification for the entire REST surface. Feed it into Postman, Insomnia, or any OpenAPI-aware client generator and get a working client without hand-typing a single request.

Where to find it

The spec is served alongside the API at:


https://api.nebbos.ai/api/v1/openapi.json

It reflects the currently deployed version: bumps land the moment the deploy pipeline promotes a build, so the spec on your machine is never more than one deploy behind production.

The spec covers all 1,652 endpoints under the /api/v1 base. Every endpoint carries its full schema, per-tier authentication requirements, error shapes, and response examples derived from the platform’s own tests.

Postman & Insomnia

Postman and Insomnia both import OpenAPI 3.x directly. Point them at the spec URL above and they will build a request collection matching the current surface, one folder per endpoint category.

For a Bearer token, import your operator token into the collection-level authorization; every request then uses it automatically. Do not paste tokens into individual requests — the collection-level pattern keeps the token in one place, and rotates cleanly.

Client generation

openapi-generator and openapi-typescript-codegen both consume the spec and produce typed clients in the language of your choice. This is the path for a language not covered by the official SDKs — the generated client is less polished than a hand-crafted one, but it is always in lockstep with the platform because it derives from the same spec.


# Example: generate a Go client
openapi-generator-cli generate \
  -i https://api.nebbos.ai/api/v1/openapi.json \
  -g go \
  -o ./nebbos-go

Contract stability

Additions to the spec (new endpoints, new optional fields) are safe within a version prefix. Removals or shape changes ship as a new version prefix, and the previous version continues to serve for at least twelve months. See Versioning & changelog for the full contract.