Search · content · answers · research · monitors

Documentation

Five services behind one bearer key: live web search, clean content extraction, grounded answers, deep research and standing monitors. Start with a concept guide for what spans every service, or go straight to the reference for the one you're calling. No account needed to read any of it.

Make your first call API reference Connect an agent

# one call, cited answer back
curl -X POST https://api.stelq.com/v1/answers/ask \
  -H "Authorization: Bearer $STELQ_KEY" \
  -d '{"query":"EU refiner exposure to Hormuz"}'

For agents

Reading this as an agent?

Every format below is generated from the same model this site renders, needs no key to fetch, and is always current. Point your agent at one of them instead of scraping these pages.

/api/docs · llms.txt index /api/docs/llms-full.txt · everything, one file /api/docs/mcp · MCP server config /api/docs/search · Search spec /api/docs/content · Content spec /api/docs/research · Research spec /api/docs/answers · Answers spec /api/docs/monitors · Monitors spec

Start here

The cross-service on-ramp: how STELQ fits together, how auth works on every call, and the one path from a fresh key to your first 200. Each service still keeps its own Quickstart on its Docs tab — these tie the five services, MCP, and the wallet into one picture.

Platform overview3 minOne bearer key, five services, one hosted MCP server, prepaid pay-per-call. The 60-second map of how STELQ fits together before you drill into any single reference — the orientation a per-service Quickstart structurally cannot give you.Read guide Authentication & API keys4 minEvery /v1/* call carries Authorization: Bearer $STELQ_KEY. Keys are stelq_live_ / stelq_test_, the secret is shown exactly once, and you pause, revoke, or expire them in the console. This is the cross-service article the per-service Auth section only summarizes.Read guide Your first call in 60 seconds3 minSet $STELQ_KEY, POST to /v1/search/query, read X-Request-Id — your first successful call is the activation moment. Signup already minted a live key and seeded $5.00 of free credit, so you reach a billed 200 before ever touching Stripe.Read guide

Money & limits

How the prepaid wallet works account-wide: what a call costs, charge-on-success-only, what a 402 means, and how to cap spend per key or auto-reload before you run dry. Spans every billed service — every claim maps to the gateway code.

Billing & credits4 minA prepaid wallet, no subscriptions: every successful call debits a fixed per-service cost, failed calls cost nothing, and you top up in whole dollars via Stripe Checkout. The cross-service money model that lives on no single service tab.Read guide Spend controls & auto-reload5 minCap any key with a per-key credit limit that auto-resets and hard-stops with no overshoot, and optionally auto-top-up off-session before you run dry. The production controls for handing a key to an agent without fear.Read guide Errors & status codes5 minEvery status the STELQ gateway actually returns — 400, 401, 402, 403, 404, 405, 409, 502, 503 — all in one uniform {"error": string} shape, with which are safe to retry and which are fatal until you fix something. This is the cross-service error contract; each service's own Errors section is the same table, filtered to that route.Read guide

Connect an agent

STELQ's wedge: one hosted MCP server exposes all five services as native tools, and a copy-ready work-order prompt hands a build order straight to a coding agent. One MCP config covers everything — so this is platform-level, not per-service.

Connect your agent (MCP & the work-order prompt)5 minOne hosted, stateless MCP server at https://mcp.stelq.com/v1/mcp exposes all five services as eleven native tools — the bearer key IS the session. Or hand a coding agent the copy-ready work-order prompt that wires itself and verifies with one real call.Read guide

Async research (the hard part)

Research is the only async, idempotent service — and where the real integration bugs live. Two guides: the submit → poll lifecycle with idempotency, and webhook signature verification (shared with Monitors). None of it applies to the three synchronous services.

Async research: submit, poll & idempotency5 minResearch is the only async service: POST to submit (charged $0.08 once), then poll GET /v1/research/run/{id} for free until completed — polling is the source of truth. Send an Idempotency-Key and a retry never double-charges; even without one, naive retries inside 90s auto-dedupe. Research-only, on purpose.Read guide Webhooks (Research completion · Monitor checks)6 minOptional signed callbacks for the two services that produce results after the call returns: Research (one terminal event per job) and Monitors (one event per check that clears your threshold). Same HMAC-SHA256 claim-check envelope pointing at result_url, same account secret, same at-least-once delivery with up to 8 retries over ~24h. The one page where getting signature verification wrong fails silently.Read guide

Standing watches

Monitors keep running after the call returns: compiled once, checked on a cadence, every check recorded and badged. The only service priced on two axes (size × cadence) and the second webhook-bearing one. One guide covers sizing, events, thresholds and pause/resume.

Monitors: standing watches, sizing and price7 minThe one service that keeps running after the call returns. How a monitor is compiled, what a check is, how size (blocks of 12 queries) × cadence sets the price, what the events feed contains, and how webhooks and pause/resume behave — every number here is the number the sweep bills.Read guide

Service references 5

One reference per live service: parameters, request and response shapes, errors, rate limits, plus a copy-ready work order for a coding agent.

SearchPOST/v1/search/queryLive web search on every request — never cached, never stale.$0.001 / querysync ContentPOST/v1/content/extractClean structured extraction — markdown + JSON, not HTML soup.$0.002 / pagesync ResearchPOST/v1/research/runDeep multi-source research with every claim cited. Quick $0.08 · Balanced $0.25 · Deep $0.40 — shown before it runs.from $0.08 / job · by depthasync AnswersPOST/v1/answers/askGrounded answers with citations, tuned for the agent loop.$0.002 / answersync MonitorsPOST/v1/monitors/createStanding watches on any topic — compiled sources, checks on a cadence, every change recorded and badged. Webhooks on what matters.$0.05 / monitor + $0.01 / checkstanding