n8n-nodes-vera

n8n community node for Vera, the web data and search API from Waggle.
Read any public page as clean Markdown, search the web, news and video, pull contacts and product data
out of a page, read full search results pages on five engines, get keyword ideas, see whether AI
assistants name your brand, run a GEO citation radar and audit a company's search footprint, all from
inside an n8n workflow.

New to Vera? Sign up for free credits. No card needed.

• Install
• Authentication
• Operations and credit costs
• Examples
• Output
• Errors
• Development
• Publishing (maintainers)

Install

In a self-hosted n8n: Settings > Community Nodes > Install, enter n8n-nodes-vera, and accept
the community-node notice. Or install it into your n8n user folder:

cd ~/.n8n/nodes && npm install n8n-nodes-vera

On n8n Cloud, the Vera node appears in the nodes panel once the package has passed n8n's
verification. See the community nodes guide.

Authentication

1. Create a Vera account (free credits on sign-up) and confirm your
   email.
2. In your Vera workspace, open API keys and create a key. It looks like vera_live_.... Copy it
   when it appears; Vera stores only a hash, so it cannot be shown again.
3. In n8n, create a Vera API credential and paste the key.

Requests go to the direct Vera API at https://vera.waggle.ae/v1 with
Authorization: Bearer <key>. The credential has a Base URL field that holds this address; leave it
as it is unless you were given a different one (for example a staging Vera). The key is sent to that
address, so only https is accepted (plain http only for localhost). The website's
/api/vera proxy is not used: it does not carry the newer routes and gives up on a long request after
40 seconds.

The credential's Test button calls GET /v1/usage, which needs a valid key and costs no credits.

Operations and credit costs

Pick a Resource, then an Operation. A failed call is not charged.

A search costs 3 credits whatever the Limit (1 to 100).

"By result" means priced by what comes back. Vera works out the most a call can cost and holds that
amount first, you are charged for what was actually served, and the rest is given back. The charge is
in credits_used on every item (and in the X-Waggle-Credits-Used response header). A results page of
up to ten results is typically 3 credits live and 1 when served from Vera's cache; every further ten
results is another page, and the Local Pack and Shopping add-ons are priced on their own. AI answers are
priced per call to each assistant and cost far more than a search, Claude the most, so start with one
assistant and one sample. On free credits a daily pace also applies to the amount held, so one expensive
call can use up a whole day: a second one the same day answers 429 until midnight UTC.

Scrape Page

Extract Contacts, Extract Products

Both take a URL. Extract Products also has a Marketplace choice: Auto-Detect (any store that
publishes Product data in its page markup, which works reliably) or Amazon, Walmart or Shopee. The
marketplace readers are best-effort: large marketplaces often block automated requests, so they may
fail, and a failed request costs nothing. They need a full HTTPS product detail URL. If the page
publishes no Product data the call is refused and not charged.

Web, News and Video Search

Query (required, up to 500 characters; quoted phrases and site: are kept as written) and Limit.
Options: Country and Language (two lowercase letters, for example ae, ar) and Idempotency Key.

SERP: Get Results Page

Keyword: Get Ideas

AI Answer: Check Brand in AI Answers

This call can take up to about 150 seconds. The node waits up to 175 seconds for it (the other search
tools wait 50 to 80 seconds, page and search calls 60).

GEO Radar: Run Query Radar

Query (required), Brand, Domain and Depth (organic results read, 1 to 100, default 30).
Options: Country, Language, Location, Skip Cache, Idempotency Key. Use it to
see how likely AI answers are to cite your brand for one topic, which forum threads to join and which
list articles name your competitors.

Company: Get Search Footprint

Company (required), Domain and Depth (1 to 100, default 20). Options: Country,
Language, Location, Skip Cache, Idempotency Key. It separates the company's own web
properties, its social and reference profiles (Wikipedia, LinkedIn, Instagram, Crunchbase), its map
listings, what third parties say and the questions people ask about the brand.

Defaults for the radar, the footprint and AI answers. If you leave Location and Country out,
the API currently uses Dubai and ae. Set them for any other market.

Idempotency Key

Set it on any operation to protect a retried call. Repeating the same key and request within 24 hours
returns the first result without charging again (the item then carries idempotency_replayed: true).
A failed call is replayed too, so retry a failed call with a new key. 16 to 128 letters, digits,
underscores, dots, colons or hyphens. Build one per item, for example
workflow-run-{{ $execution.id }}-item-{{ $itemIndex }}.

Examples

Rank check with local listings (SERP > Get Results Page): Engine Google, Query coffee shop,
Add-Ons Local Pack, Options > Location Dubai, Country ae. Each item has data.organic (ranked
results) and data.serp_features.local_pack (the map listings).

Keyword research (Keyword > Get Ideas): Seed specialty coffee, Expand Questions, Options >
Country ae. Add a Split Out node on data.ideas to get one item per idea, each with keyword,
sources and intent.

Is my brand in the answer? (AI Answer > Check Brand in AI Answers): Question What are the best specialty coffee roasters in Dubai?, Brand Roasters, Brand Domain roasters.ae, Competitors
Other Roasters, Assistants ChatGPT. Read data.engines[0].aggregate for the mention rate and share
of voice, and data.engines[0].samples[].citations for the pages the assistant cited. Schedule it
daily to track visibility.

Output

One item per input item, with the API's JSON plus the credits used:

{
  "request_id": "req_...",
  "data": {
    "url": "https://example.com/",
    "title": "Example Domain",
    "markdown": "# Example Domain\n...",
    "text": "...",
    "links": [],
    "metadata": {},
    "extracted": { "price": "$19" }
  },
  "credits_used": 1,
  "elapsed_ms": 412
}

credits_used is what that call was charged (from the X-Waggle-Credits-Used header; 0 on an
idempotent replay). Searches return rows under data.results; use a Split Out node on that field to
get one item per result. Contacts come back under data.emails, data.phones and data.socials,
products under data.products.

The search tools return the search service's JSON under data, unchanged:

Errors

The node raises an error with advice you can act on. Turn on Continue on Fail to get an error item
instead (error, description, status_code, and retry_after_seconds or request_id when known).

Development

npm install
npm run lint        # n8n's eslint rules, including the n8n Cloud eligibility checks
npm run build       # tsc plus static files, into dist/
npm test            # vitest, HTTP mocked
npm run typecheck   # type-checks the source and the tests
npm run dev         # n8n with this node loaded (needs Docker or Podman; see the n8n docs)

The tests mock n8n's HTTP helper, so they never reach the network or spend credits. A contract test
(test/contract.test.ts) checks every request body this node builds against a snapshot of Vera's
published OpenAPI request schemas and example bodies (test/fixtures), refreshed from
https://vera.waggle.ae/v1/openapi.json.

How the fields map to the API

Publishing (maintainers)

The repository is https://github.com/Studio-Hyve/n8n-nodes-vera. npm provenance refuses a package
whose repository.url does not match the repository that builds it, so keep package.json pointing
there.

Verified node (needed for n8n Cloud)

Since 1 May 2026 n8n only verifies community nodes published from a GitHub Actions workflow with an npm
provenance statement, not from a laptop. This repository ships that workflow
(.github/workflows/publish.yml).

1. Create an empty public GitHub repository named n8n-nodes-vera in the Studio-Hyve organisation
   (no README, licence or .gitignore), then push this one to it:

   git remote add origin git@github.com:Studio-Hyve/n8n-nodes-vera.git
   git push -u origin main
   

2. Give the workflow permission to publish, either of:

   • Trusted Publisher (no stored secret): on npmjs.com, open the package's settings, then
     Trusted Publishers > GitHub Actions, with repository owner Studio-Hyve, repository
     n8n-nodes-vera and workflow publish.yml. (npm lets you add it once the package exists, so if
     npm will not accept it before a first release, use the token below for that first release.)
   • Token: on npmjs.com create a Granular Access Token with read and write access to this package
     and save it as the GitHub Actions secret NPM_TOKEN.

3. Release. This lints, builds, asks for the version bump, updates CHANGELOG.md, commits, tags and
   pushes; the pushed tag triggers publish.yml, which publishes with provenance:

   npm login
   npm run release
   

4. Submit the published package for verification in the
   n8n Creator Portal. n8n checks that npm run lint passes, that
   there are no runtime dependencies, and the UX and documentation guidelines.

Quick publish (self-hosted n8n only, not eligible for verification)

prepublishOnly blocks a bare npm publish on purpose, to steer you to npm run release. To publish
once from your machine anyway:

npm login
npm run lint && npm run build && npm test
npm publish --ignore-scripts          # add --otp=<code> if your npm account uses 2FA

Anyone can then install it with Settings > Community Nodes on a self-hosted n8n. n8n will not verify
a package published this way, so it will not reach n8n Cloud.

Licence

MIT