Documentation
MCP tool reference

Start

OverviewQuickstartAuthenticationMaking callsDiscoveryMCPErrors & billingAPI schemaPay with x402

Guides

Web searchOneSearchOneFindOneScrapeResearchWeb scrapingGTM & leads

Data APIs

Public recordsReal estateSocial dataOn-chain & marketsCompliance & KYBOpen data

Execution APIs

BrowserLLMsMediaComputeFiles & storageEmailCommerce

vaaya / docs / reference

Vaaya API reference

Vaaya gives you access to data, LLMs, media, search, scraping, compute, and storage services through a single API. It holds the vendor accounts and settles with each provider for you. You pay per call, and only when the call succeeds. Start with the Quickstart.

One call shape

Every catalog action uses the same endpoint: POST /api/run/{service}/{action}. The request body contains the action’s inputs. Your API key goes in the Authorization header. Every response uses the same structure.

curl -X POST https://vaaya.ai/api/run/exa/search \
  -H "Authorization: Bearer $VAAYA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "latest developments in WebTransport", "numResults": 10}'

If you don’t know which service you need, POST /api/run/vaaya/consult (1¢) turns a plain-English request into exact calls. POST /api/run/vaaya/discover (free) searches the available endpoints.

The services

Each service page explains what the service does, what each action costs, and which vendor to pick.

CategoryWhat it covers
Web searchSemantic, keyword, and cited-answer search across five vendors, from 1¢ per query.
OneSearchSearch multiple sources and compare the results in one fixed-price call.
OneFindPeople as rows from a plain-English query. Find for 2¢, add contact data per filled field.
OneScrapeRead a list of URLs into structured rows. Standard retrieval costs 2¢ per URL; unblocking is billed per page read.
Web scrapingContent extraction, JS rendering, crawls, structured extraction, and anti-bot escalation.
BrowserCloud browser sessions for clicking, typing, and form-filling.
LLMs340+ models behind one OpenAI-compatible endpoint, metered per token.
MediaImage, video, audio, and speech generation. The model is a parameter.
ComputeEphemeral sandboxes and GPU boxes for running code in isolation.
Files & storageDurable file storage your agent can write and share.
EmailInboxes your agent owns. Send, receive, and act on replies.
CommerceReal-world actions: faxes, printed mail, purchases, hiring other agents.
GTM & leadsFind, enrich, and verify prospects. Research prospects and prepare outreach.
ResearchManaged multi-step research runs that produce cited reports.
Public recordsSEC filings, court dockets, nonprofit 990s, and H-1B salaries.
Real estateUS property records, valuations, listings, and owner data.
Social dataProfiles, posts, and comments across social platforms, including CN platforms.
On-chain & marketsCrypto, wallet forensics, and prediction-market data.
Compliance & KYBSanctions, PEP, adverse media, and company-registry checks.
Open dataSearch the catalog for public datasets and specialist APIs.

Pricing model

  • Every action has a listed price in cents. It appears on the service page and in the machine-readable index.
  • You are only charged when a call succeeds. An upstream error charges you nothing.
  • Pass max_cost_cents to set a ceiling on any call. The response reports the actual charged_cents.
  • Personal accounts start with $2 welcome credit; Instinct accounts get $10, subject to signup checks. No card required. Every call is itemized at /balance.
  • Documented prices are defaults or ranges. Check GET /api/run/{service}/{action} before relying on a price in production.