Skip to main content

Provider and data-source map

This map lists external providers, APIs, datasets, and platforms used or planned across the inspected product sources. It intentionally excludes secret values, account IDs, tokens, and credentials.

Cloud and platform providers

Provider/APIUsed byPurposeAuth/secrets implicationsReuse opportunityDuplication risk
Cloudflare WorkersSiteIntel, Trackpal, DataStitcher, BigQuery Atlas, Bollocks Radar, Product Compass, Meridian HubPublic/admin apps, docs sites, workers, progress workerWorker secrets and Access policies need status-only handlingShared deployable-unit docs, admin app patterns, secret-status componentsRepeated OpenNext/Worker setup and Access rules
Cloudflare PagesBigQuery AtlasPublic catalogue hostingBuild/deploy config, no browser secretsLow-cost static/public catalogue patternSeparate Pages/Workers conventions
Cloudflare D1BigQuery Atlas, Meridian HubRead-side cache and Hub databaseBinding IDs should not be treated as secrets but should stay in configD1 read-cache patternSchema drift across apps
Cloudflare R2BigQuery Atlas, SiteIntel, Meridian HubLarge JSON/schema/raw artefact storageBucket bindings and access controlsD1 compact rows plus R2 full payload patternRaw artefact leakage if pointers are exposed badly
Cloudflare AccessTrackpal, SiteIntel, DataStitcher, BigQuery Atlas, Bollocks Radar, Meridian HubProtect admin/docs surfacesAccess app config and policy status onlyShared protected-admin blueprintInconsistent admin auth assumptions
Cloudflare Queues/CronTrackpalSync jobs and private-beta scheduled workWorker secrets for provider syncSmall sync-worker patternPremature use where Cloud Tasks is already chosen
Google Cloud RunSiteIntel, DataStitcher, BigQuery Atlas, Bollocks RadarHeavy/private workers and jobsRuntime service accounts and Secret ManagerBounded worker/job patternOne-off worker deployment scripts
Google Cloud TasksSiteIntel, Bollocks RadarBounded async task dispatchQueue invoker/auth configShared lock/retry queue patternDuplicate queue state models
Google Cloud Scheduler/WorkflowsDataStitcher, BigQuery Atlas, Bollocks RadarScheduled and orchestrated ingestionScheduler/workflow invoker accountsGeneric scheduled ingestion patternSource-specific workflows
Google Cloud StorageDataStitcher, Bollocks Radar, SiteIntel ideas/current compatible contractsRaw, parsed, manifest, and profile artefactsBucket IAM and regional policyRaw artefact boundaryRegion/policy mistakes
Google BigQueryDataStitcher, BigQuery Atlas, SiteIntel, Meridian packagesWarehouse, public metadata, domain registry, martsGoogle identity, dataset IAM, query cost controlsBigQuery catalogue package/patternUnbounded queries or duplicated schema code
Supabase/Postgres/NeonSiteIntel, Trackpal, DataStitcher, Bollocks Radar, BigQuery Atlas optionalControl-plane, registry, state, metadataDirect DB URLs and service roles must stay server-sideShared control-plane run/source schemasExposing private ops schemas through public APIs

Product data sources and APIs

Provider/APIUsed byPurposeAuth/secrets implicationsReuse opportunityNotes
Google Fact Check Tools APIBollocks RadarClaim and fact-check review ingestionAPI key server-side onlySource adapter patternSearch API, not complete feed
GDELT DOC 2.0Bollocks Radar plannedTrend/context enrichmentAPI/key terms to verifyEnrichment adapterNot MVP source
YouTube Data APIBollocks Radar plannedMedia/video contextAPI key and quotaEnrichment adapterRequires abuse controls
Wikidata SPARQLBollocks Radar plannedEntity/context enrichmentUsually no secret, rate limits applyPublic reference adapterWatch query limits
Google Cloud Vision Web DetectionBollocks Radar laterMedia verificationGCP credentials, costMedia-source patternRequires careful confidence wording
GitHub Code SearchBigQuery AtlasBigQuery identifier discoveryPAT or GitHub tokenEvidence source adapterProvider-specific rate limits
Brave Search APIBigQuery AtlasCandidate discoveryAPI keySearch provider adapterPaid/limited provider
Google Programmable Search JSON APIBigQuery AtlasCandidate discoveryAPI key and CXSearch provider adapterSource docs note new customer limitation
DataForSEOBigQuery Atlas, SiteIntel, Bollocks Radar plannedSERP, dataset, trends, SEO/competitor dataLogin/password/API credentialsPaid provider adapter and cost controlsMust be disabled by default unless configured
Common Crawl CDXBigQuery AtlasDiscover URLs likely to mention BigQuery docs/pagesNo secretPublic web evidence adapterNeed politeness/caps
Stack Exchange APIBigQuery AtlasSQL/error examples with BigQuery IDsOptional keyEvidence adapterRate limits
OpenAlexBigQuery AtlasResearch/publication pages mentioning datasetsNo secretEvidence adapterPublic API terms
BigQuery metadata APIsBigQuery Atlas, SiteIntel, DataStitcherDataset/table/schema metadataGoogle identity and quotaStrong candidate for shared helperMetadata-only still needs bounds
data.gov.au CKANDataStitcherPublic dataset discoveryUsually no secretCKAN connectorSource-specific pagination and formats
ABR/ATO/ABS/ASIC/ACNC and other official registersDataStitcherBusiness/public audit datasetsVaries; often public bulk feedsSource registry plus adaptersWording and publication policy matter
DNS/RDAP/TLS/HTTP/robots/sitemap/homepageSiteIntelDomain report source tasksMostly no secrets, but SSRF/abuse controlsTask adaptersMust cap time/size/redirects
Google PageSpeed InsightsSiteIntelPerformance checksAPI key may be useful/required for quotaSource task adapterCost/quota controls
Cloudflare Radar/Intel APIsSiteIntelDomain rank/category/risk signalsCloudflare token/account configProvider adapterRequires careful result wording
Mozilla Observatory / SSL Labs / OpenPageRankSiteIntelSecurity/authority signalsOpenPageRank requires keySource adaptersRate limits and data caveats
Google AdsTrackpalRead-only spend/performance metricsOAuth, developer token, approvalConnector interfaceVerify current docs/scopes before work
Meta AdsTrackpal planned/stubbedRead-only spend/performance metricsApp review and OAuthConnector interfaceGuarded stub
ShopifyTrackpalStore pulse metricsOAuth/admin scopesConnector interfaceRevenue definitions need caveats
KlaviyoTrackpalEmail revenue/share metricsAPI credentials/scopesConnector interfaceAttribution windows matter
XeroTrackpal planned/stubbedP&L/cost mappingOAuth/accounting scopesConnector interfaceLater phase
models.dev pricing APICodex Sessions AnalyserOptional model pricing estimatesNetwork fetch or saved file; no raw sessionsPricing cache helperTime-sensitive external data

Shared adapter recommendations

  1. Build shared contracts before shared provider code.
  2. Keep provider-specific auth, pagination, quotas, terms, and billing visible in docs.
  3. Use status-only secret/config surfaces in admin apps.
  4. Do not expose ops schemas, service roles, or raw provider payloads to browser-readable routes.
  5. Add current official-doc verification tasks before implementing live ad/ecommerce/accounting connectors.

Sources consulted

  • Product dossiers under docs/projects/*/overview.md
  • projects/site-intel/docs/architecture/data-source-registry.mdx
  • projects/trackpal/docs/product/12-integrations-strategy.md
  • projects/bq-atlas/services/candidate-crawler/README.md
  • projects/data-stitcher/docs/public-audit-sources.md
  • projects/bollocks-radar/docs/reference/technical/google-fact-check-ingestion.mdx