Enable optional web search

Slug guessing cannot derive every board identity: a Workday slug such as autodesk.wd1.Ext contains tenant, pod and site. Configure a SerpApi key to allow searchForBoards and resolution’s fallback search.

import { configureSearch, resolveCompany } from '@chrisdothtml/job-scraper';

configureSearch({
  apiKey: 'your SerpApi key',
  limit: 100,
  period: 'month',
});

await resolveCompany('Whatnot');
// { name: 'Whatnot', scraper: 'WorkdayScraper', slug: 'whatnot.wd1.Whatnot', discovered: true }

Settings and defaults

configureSearch merges partial overrides. undefined removes an override; getSearchConfig shows effective settings.

Option	Default	Purpose
apiKey	Node environment, otherwise null	No search without a key
limit	250	Searches allowed per period
period	'month'	'day' or 'month' quota window
resultCount	10	Results requested for an individual search
resultTtl	30 days in milliseconds	Reuse remembered board results, including misses
batchSize	5	Companies per batched query
homepages	false	Enable homepage search by default

Node can supply:

SERP_API_TOKEN=... # alternatively SERPAPI_KEY or SERPAPI_API_KEY
SERPAPI_SEARCH_LIMIT=100
SERPAPI_SEARCH_PERIOD=month
SERPAPI_HOMEPAGE_SEARCH=1

Browsers have no environment-variable fallback. Choose quota settings from your own provider plan; the package defaults are not a promise of provider allowance.

Control cost

Usage is persisted in search-usage.json and resets when the UTC day/month rolls over. A query is counted before its response arrives: a failed request can still consume provider quota. Once the limit is reached, new queries stop and resolution relies on earlier steps. canSearch reports key/quota availability.

Remembered board results live in search-results.json, including fruitless searches. Equivalent concurrent work shares an in-flight request. searchForBoardsBatch skips remembered companies, ORs up to batchSize names into each query, asks for up to 100 results, and attributes matching slugs back to names:

import {
  clearSearchCache,
  searchForBoardsBatch,
} from '@chrisdothtml/job-scraper';

await searchForBoardsBatch(['Klaviyo', 'Notion', 'Whatnot']); // one search
await clearSearchCache(); // start over

A result is a candidate, not proof. Resolution checks both a plausible company-name slug match and the same minJobs activity threshold as guessing before registration. A stale or unrelated search hit must not become a permanently failing registry record.

Disable search for one lookup

import { resolveCompany } from '@chrisdothtml/job-scraper';

await resolveCompany({ company: 'Whatnot', search: false });

clearSearchCache clears remembered board results, not quota usage. searchForHomepage returns unverified website candidates; use resolveHomepage for checked, remembered homepage results.