Package Information
Available Nodes
Documentation generating…
Documentation
n8n-nodes-banneroven
An n8n community node for BannerOven: render images, render packs, make signed image links, and read
templates, usage, jobs and batches, from an n8n workflow. No runtime dependencies.
Install it in n8n under Settings > Community nodes > Install with the package name n8n-nodes-banneroven, or in a
self-hosted n8n from the command line:
npm install n8n-nodes-banneroven
The package is 0.x: its parameters may still change between minor versions. It is named after n8n's ownn8n-nodes-* convention for community nodes rather than the @banneroven/* scope (decision D-6.6.1).
Credential: BannerOven API
- API Key — from your workspace's API keys page. Give this node a key scoped to
render,readandsign
only; it never needswriteorbilling(seedocs/key-scopes.md). - Base URL — the address of your BannerOven instance. Must be
https;httpis accepted only forlocalhost,127.0.0.1and[::1]. This is the one source of the address the node calls — see "Base URL
override" below for the only other way.
Use the credential's Test button to check it: a 200 passes, any 401 or 403 fails with the server's own
message.
Resources and operations
| Resource | Operation | Calls |
|---|---|---|
| Image | Render | POST /api/images, answered as binary data in the data output property (25 MB cap) |
| Image | Pack | POST /api/images/pack, answered as JSON |
| Image | Sign URL | POST /api/images/sign, answered as JSON |
| Template | List | GET /api/templates |
| Template | Describe | GET /api/templates/{id} |
| Usage | Get | GET /api/usage |
| Job | Get | GET /api/jobs/{id} |
| Batch | Create | POST /api/batches (execute: true) |
| Batch | Get | GET /api/batches/{id} |
Field coverage is not exhaustive of every option docs/openapi.yaml names: the common ones are exposed
directly, and Content, Defaults, Items and Metadata stay free-form JSON fields so anything else reaches
the request without a field of its own.
Options
- Max renders per execution (default 100) refuses to make more than this many calls in one execution,
checked before the first call. This is the node's own bound on a workflow loop's spend; your workspace's own
rate limits are the other half of that defense (see "Security" below). - Base URL override (under "Options") is the one, explicit, separate way to call an address other than the
credential's for one call. It is never read from an upstream item's JSON or binary data on its own — you type
or wire it here on purpose.
Idempotency
Every call n8n's own retry-on-fail would retry carries an Idempotency-Key derived from the item's own
resource, operation and parameters. A retried attempt of the same logical item reuses the same key (so it is
never charged twice); a different item gets a different key.
Security
- SSRF (baseUrl). The address comes from the credential only;
httpsis enforced (loopback excepted) at
call time. A per-call override is a separate, explicitly-labelled option, never silently read from an
upstream node's data. - Unbounded spend. "Max renders per execution" bounds one run of this node; it does not replace your
workspace's own plan limits and rate limits. If a public form can trigger the workflow this node sits in,
treat that form the way you would treat a public API key: a dedicated low-limit key, test mode first, and this
option set low. - Credential leakage. The API key is never written into an output item or logged;
test/render.test.ts
checks this for the Render operation. - Content-derived file names. The binary file name is always
banner-<generation id>.<ext>, using the id
the server returned — never a template name or slot content.
Coverage
scripts/check-n8n-coverage.ts (run from the repository root as pnpm check:n8n) fails when a read or render
operation of docs/openapi.yaml is neither implemented here nor listed, with a reason, inoperation-map.json.