The live picture, free. Depth, metered.

All-in-one bundle: verdict, live oil prices, IMF PortWatch daily transit count + throughput, AIS concurrent presence overlay, insurance multiple, trade impact, carrier suspensions, pipeline bypass, SPR, gas, and a per-section dataHealth chip. Powers the embed widget; recommended starting point for newsroom integrations. marketTime: the trade time the upstream reported for the Brent quote; oilAsOf is when we last checked.

curl https://straits.live/status

{
  "asOf": "2026-10-16T14:41:06Z",
  "status": "closed",
  "verdict": { "status": "closed", "short": "Effectively no", "long": "…", "basis": "insurance" },
  "brent": 105.28, "wti": 97.32, "change24h": 0.27,
  "transits": { "count": 2, "baseline": 85, "throughputPct": 2, "asOfDate": "2026-10-11", "asOfAgeDays": 5, "stale": false },
  "insurance": { "multiple": 30, "vlccPremiumLow": 250000, "vlccPremiumHigh": 10000000 },
  "hormuzIndex": { "crisisPressure": { "value": 87 }, "escalationProbability": { "value": 60 } },
  "carrierSuspensions": [ … ],
  "dataHealth": { "ships": { "source": "live" }, "oil": { "source": "live" }, … }
}

Crisis Pressure and Escalation Forecast composites: current reading plus a per-component breakdown for each. Time series is opt-in: append ?history=24h, 7d, or 30d, returned nested at indices.<name>.history (30d resolves to daily points from the historical rollup, the rest to the raw series, 10 minutes apart (5 before 2026-11-03)).

Queryhistory (24h | 7d | 30d)

curl https://straits.live/api/index

{
  "asOf": "2026-08-23T13:00:00Z",
  "indices": {
    "crisisPressure": { "value": 94, "band": "extreme", "components": { … } },
    "escalationProbability": { "value": 56, "band": "elevated", "components": { … } }
  }
}

Rolling Brent and WTI prices from intraday front-month futures contracts (Yahoo Finance chart data, continuous quotes as fallback, EIA daily backstop). Companion to /data/oil.csv for programmatic consumers. The points are futures. The latest block normally holds a futures print too, but when the futures feed has been stale for over 2 hours it holds the EIA daily Dated Brent spot close, which can sit well off futures. Tell them apart with latest.sourceKind ("eia" for the spot close, null for futures) and latest.observedAt (the trading day of that close). source and stale are unchanged by it. latest.marketTime: the trade time the upstream reported for the Brent quote; asOf is when we last checked.

Querylimit (the newest N stored points, default 168, max 2000; points are not evenly spaced, so N points cover an irregular span, not a fixed number of hours), since (ISO timestamp; use it for a time window, and pair it with a large limit when the window is long)

curl 'https://straits.live/api/v1/oil?limit=2'

{
  "asOf": "2026-06-10T17:23:00Z",
  "count": 2,
  "pointCount": 2,
  "source": "Intraday Brent / WTI front-month futures contracts …",
  "points": [
    { "t": 1781110302601, "iso": "2026-06-10T16:51:42Z", "brent": 93.99, "wti": 91.06 },
    { "t": 1781112144785, "iso": "2026-06-10T17:22:24Z", "brent": 94.12, "wti": 91.20 }
  ]
}

Top Hormuz / Iran-conflict / oil-shock contracts from Kalshi (api.elections.kalshi.com), Polymarket (gamma-api.polymarket.com) and Manifold. A keyword filter keeps out contracts that do not name Hormuz, Iran or oil; election-framed events that do name Hormuz can appear, and the Hormuz Index excludes them (see /methodology). Each row carries excludedFromIndex, true when the index drops it, and Kalshi KXHORMUZWEEKLY rows carry ladderComplete: false because only the top three rungs by volume are stored. liquidityUsd is null where a venue does not report it (Kalshi). Kalshi prices are the last trade, held inside the current bid and ask, when the rung traded in the last 24 hours, and otherwise the midpoint of a two-sided quote (priceSource: trade or mid); bid, ask and volume24h let you judge how thin a rung is, and are null on the other venues along with priceSource. total counts the rows that passed the deadline and venue filters before the limit, limit is the value applied, and truncated is true when total is more than count.

Querylimit (default 50, max 50), venue (kalshi | polymarket | manifold)

curl 'https://straits.live/api/v1/markets?venue=polymarket&limit=1'

{
  "asOf": "2026-06-10T17:23:00Z",
  "count": 1,
  "total": 15,
  "limit": 1,
  "truncated": true,
  "markets": [
    { "id": "polymarket_665374", "venue": "polymarket",
      "title": "Will the U.S. invade Iran before 2027?",
      "probability": 0.175, "volumeUsd": 35039246.83,
      "priceSource": null, "bid": null, "ask": null, "volume24h": null,
      "closeAt": "2026-12-31T00:00:00Z",
      "sourceUrl": "https://polymarket.com/…" }
  ]
}

Per-state regular-grade averages plus a national mean computed across the 50 states + DC.

Querystates (1 / true to include the per-state list; default omitted for payload size)

The dashboard's authoritative throughput figure. Carries the per-type breakdown (tanker, cargo, container, dry bulk, general cargo, Ro-Ro), capacity in deadweight tonnage, the trailing 30-day rolling baseline median, and a separate pre-crisis baseline median (fixed window: the final pre-crisis year, 2025-02-28 to 2026-02-27) used as the stable peacetime anchor. The per-type fields nest: nTotal is nTanker plus nCargo, and nCargo is nContainer plus nDryBulk plus nGeneralCargo plus nRoro, so summing all seven fields double-counts. Not a programmer? The same series is a readable day-by-day table at /how-many-ships-passed-through-the-strait-of-hormuz-today.

Queryhistory (1 / true to include chokepointTransitsHistory rows), limit (default 90, max 365: the newest days only; /data/transits.csv has the full series from 2025-02-28)

curl https://straits.live/api/v1/transits

{
  "asOf": "2026-08-30",
  "generatedAt": "2026-09-09T13:02:00Z",
  "methodology": "IMF PortWatch daily vessel transit count …",
  "latest": {
    "date": "2026-08-30",
    "nTotal": 6, "nTanker": 2, "nCargo": 4, "nContainer": 0,
    "capacity": 70732,
    "baselineMedian": 5,
    "preCrisisBaselineMedian": 85,
    "ma7": 5.57,
    "provisional": false
  }
}

Side-by-side daily counts for Hormuz, Bab el-Mandeb, Suez, the Cape of Good Hope, and Malacca, sorted Hormuz first, then by nTotal descending, largest route by vessel count first.

Queryhistory (1 / true to include per-chokepoint daily history[])

The upstream port-page aggregates this fed went behind a subscription on 2026-08-28. The endpoint still responds but returns a frozen snapshot from that date, marked "RETIRED FEED, DO NOT READ AS CURRENT". Do not poll this as an AIS fallback; for day-by-day transit counts use /api/v1/transits or the readable table at /how-many-ships-passed-through-the-strait-of-hormuz-today.

Same upstream loss as /api/v1/hormuz-snapshot: the source port-page aggregates went behind a subscription on 2026-08-28. The endpoint still responds but returns a frozen snapshot from that date, marked "RETIRED FEED, DO NOT READ AS CURRENT".

Live counts at Fujairah, Sohar, Bandar Abbas, and Jebel Ali derived from live AIS positions inside a 10–18 km radius per port. Broken down by status (anchored, transiting) and vessel type.

Anchored-vessel count inside the Persian Gulf bounding box, hourly. Powers the homepage sparkline. count (also sent as pointCount) is the number of points in the series, not a vessel count; latestAt is when the underlying feed was last refreshed, while asOf is the start of the newest hour.

Querylimit (default 168 = 7 days hourly, max 720), since (ISO)

Hourly cross-tab of vessels in transit by vessel type (tanker, cargo, military, other). Pairs with /api/v1/stranded for a full picture of the strait's vessel mix over time. The response carries a methodology field covering how type is resolved and why the military count is a floor. count (also sent as pointCount) is the number of points in the series, not a vessel count.

Querylimit (default 168 = 7 days hourly, max 720), since (ISO)

Free summary: the count of tankers gone dark inside the strait core area plus the 7-day baseline. Filters: type === tanker, totalSightings ≥ 50, last position in lat 24–28 / lng 55–58, last seen 3–24 hours ago. The per-vessel dark-fleet identities are in the premium vessel-intelligence API.

Free summary: the count of vessels visible on AIS in the Persian Gulf window whose identity matches an OFAC SDN entry (IMO / normalised-name join; false positives possible). The per-vessel manifest with sanction programs, flag, and call sign is in the premium vessel-intelligence API.

Free summary: how many vessels are on live AIS right now in the Persian Gulf window, broken down by corridor and ship type. Only the Gulf is measured: region=cape returns null counts with a note, and any other region value returns 400. The per-vessel manifest with identities is in the premium vessel-intelligence API. ?flagged=1 counts the OFAC-SDN-matched subset.

Queryflagged (1 / true for OFAC-matched only), region (hormuz; cape is accepted but not measured)

Multiple over the peace-baseline VLCC premium, per-voyage cost in USD from the peacetime baseline to the current crisis high, and the list of P&I clubs and war-risk underwriters that have publicly withdrawn cover, imposed quotas, or refused renewals.

curl https://straits.live/api/v1/insurance

{
  "asOf": "2026-08-05T17:23:00Z",
  "methodology": "War-risk insurance multiple compares …",
  "insurance": {
    "multiple": 30,
    "vlccPremiumLow": 250000,
    "vlccPremiumHigh": 10000000,
    "withdrawnClubs": ["Gard", "Skuld", "North Standard", …],
    "sourceName": "straits.live estimate · carrier advisories, Lloyd's List, TradeWinds, Reuters"
  }
}

EIA/IEA shares of global seaborne oil and LNG that transit the strait, a curated estimate of the daily economic cost, the most-affected importing economies (ranked by the share of their crude imports that crosses the strait), and the extra voyage days of the Cape reroute.

Marine war-risk high-risk-areas designations as published in the industry circular, plus a flag noting whether the latest revision mentions the Arabian Gulf. Drives war-risk premium loads at most London-market underwriters. updatedAt restamps on every check. lastChangedAt is when straits.live first saw a new circular, and circularDate is the date the committee published it.

Top global container carriers (ranked by TEU capacity) and their current posture: suspended, rerouting, limited, or transiting. Includes vessels-trapped count, combined trapped-TEU capacity, and the per-carrier surcharge in USD where a carrier notice states one (indicative editorial reading, null otherwise; surchargeUnit says whether it is per container or per TEU). hormuzPosture is the operative field: ≥5 carriers with hormuzPosture "stopped" drives the dashboard's "effectively closed" verdict, counting only rows whose advisoryDate (the carrier's own advisory date) is 60 days old or less.

curl https://straits.live/api/v1/carriers

{
  "asOf": "2026-09-15T17:01:00Z",
  "count": 9,
  "carriers": [
    { "id": "cma-cgm", "carrier": "CMA CGM", "rank": 3, "status": "rerouting",
      "surchargeUsdPerTeu": null, "surchargeUnit": null, "vesselsTrapped": 8, "trappedTeu": null,
      "notes": "…", "sourceName": "CMA CGM Q2 earnings call, via WorldCargo News", "sourceUrl": "https://…" }
  ]
}

Petroline (Saudi Arabia → Yanbu) and Habshan-Fujairah (UAE → Gulf of Oman) are the main lines that move crude out of the Gulf without transiting Hormuz. Each row carries nameplate capacity in BPD and current utilisation percent, plus utilizationSource and utilizationAsOf, which say whose rate that is and the day the outlet reported it (null where no rate is stated).

Sell/buy rates in both Toman (1 Toman = 10 Rial) and Rial, plus 24-hour and 7-day percentage deltas off the midRial. The basis field is "direct" for a quoted dollar rate or "cross-aed" when the source is not quoting the dollar and the figure is implied from the UAE dirham rate at the 3.6725 peg.

JSON archive of every brief written by the dailyBrief Cloud Function, keyed by UTC date. Each brief carries 3–6 bullets, a short summary, the day's Brent price plus 24-hour delta, and the day's transit count. source is ai-assisted, curated or mock (the template fallback); before 2026-10-02 ai-assisted briefs carried a vendor-specific value.

Querydate (YYYY-MM-DD for a single brief; 404 if missing), limit (default 30, max 120)

Bare-minimum payload polled by the homepage ticker every two minutes: Brent price, 24-hour change, transit and stranded counts, plus a verifiedAt anchor. The Brent price is a futures print, except that when the futures feed has been stale for over 2 hours it is the EIA daily Dated Brent spot close. oil.sourceKind is "eia" for that case (null otherwise) and oil.observedAt is the trading day of the close; oil.source and oil.stale are not flipped for it. oil.marketTime: the trade time the upstream reported for the Brent quote; oil.updatedAt is when we last checked.

Compares the current state to a caller-provided anchor timestamp. Returns a material flag once any of: ≥24h elapsed, |Brent %| ≥ 1, ≥1 event, ≥1 carrier advisory dated after the anchor, |transit delta| ≥ 30 hulls, or |stranded delta| (the offshore-holding count) ≥ 30 hulls (smaller AIS deltas are returned as null). Events cover only rows still held in the rolling window of up to three days, so for an older anchor the count is partial; eventsFloor is true when the anchor is over 72 hours old and marks the count as a minimum. An anchor under 6 hours old or over 90 days old returns material: false with no deltas.

Queryat (required ISO timestamp)

Backed by EIA Weekly Petroleum Status Report. Carries the 4-week change so spike-vs-baseline reads correctly.

Cushing, Oklahoma is the WTI delivery hub. Inventory swings here track the North-American crude balance independent of the Gulf-Coast / SPR picture; pairs with the Brent–WTI spread to indicate whether disruption is global or American-specific.

Indexed strikes, ship incidents, closures, and negotiations. Mirrors the curated Firestore record so consumers see the same source attribution as the homepage timeline. Companion to /data/events.csv and /feed.xml. count is headlines, one row per outlet report; stories (added 8 October 2026) groups the same rows by incident, because several outlets can report one.

Querysince (ISO), type (strike | ship_incident | closure | negotiation), limit (default 50, max 200)

curl 'https://straits.live/api/v1/events?type=strike&limit=2'

{
  "asOf": "2026-05-27T14:30:00Z",
  "count": 1, "stories": 1,
  "events": [
    { "id": "ev_2026-05-26_…", "type": "strike", "severity": "high",
      "occurredAt": "2026-05-26T11:20:00Z",
      "title": "Strike reported on Bandar Abbas refinery",
      "sourceName": "Reuters", "sourceUrl": "https://…" }
  ]
}

The live-AIS vessel manifest (MMSI-keyed, with a best-effort flag derived from the MMSI MID), the OFAC-sanctioned vessels the AIS feed lists in the Gulf (last fix up to 12 hours old, aged per row) with program detail, per-vessel risk bands, latest raw AIS positions, the top AIS-dark hulls with the full dark count alongside, and OFAC + UK sanctions reference-list sizes. Per-IMO multi-body sanction detail is in the single-vessel dossier.

Feeds: stranded and ais-gaps (hourly buckets with gaps, up to a full year; the free windows stop at 30 days and none), index-5min (both Hormuz indices at every tick, 10 minutes apart since 2026-11-03 and 5 before, with the pre-rails weighted average, ?days= 1 to 7 a page (a larger value is cut to 7, and the response says so); ?before=<ISO> with the response’s nextBefore pages back through the full tick history), and transits (daily PortWatch history since 2025-02-28 (upstream goes back to 2019), enriched per day with the trailing 7-day moving average, deviation from the pre-crisis baseline, and headroom to the reopening-resolution threshold, the analytics the free raw CSV omits). One contract for all four series.

The entire dashboard state in a single response: verdict, oil, transits, insurance, carriers, pipelines, ports, events, prediction markets, both indices with component audit, and the full daily brief.

The AI-assisted situation brief as structured JSON: prose summary, bullets, and the day’s key figures, by date or latest, with archive access. The prose itself is free on /report; this is the stable machine contract.

One vessel by IMO: the unified multi-body sanctions view (every body that lists it: US OFAC and the UK/FCDO list, with the UK row adding owner/operator, flag, tonnage, ship type, year built, and Date Designated), merged identity, per-vessel risk band, and dark-activity (AIS-gap) record. Resolves for IMOs listed by OFAC or the UK, or with a dark-activity record.

Every requested series (index-5min, stranded, ais-gaps, transits) in one archive, as JSON or multi-section CSV. stranded and ais-gaps cap at 365 days of retained history; transits returns its full retained history instead (daily since 2025-02-28 and growing), carrying the same per-day analytics (7-day MA, baseline deviation, reopening-threshold headroom) as the per-feed history call. index-5min is every tick for the last 7 days and daily open, close, min, max, mean and band before that (from the daily rollup, closed days only); its resolution, window and note fields say so, and ticks older than the 7-day window come from the per-feed history call with ?before=. ?feeds= to select, ?format=csv for the download.

POST to buy a short-lived signed token, then open a Server-Sent Events connection at /api/premium/stream?token= to receive ticks, Hormuz Index moves, and events live instead of polling. One payment buys one bounded streaming window.

POST a target URL to receive a signed webhook whenever the verdict flips, a Hormuz Index band crosses up, or a monitored threshold breaches. One payment funds 30 days or 100 deliveries, whichever comes first.

The flag states of the live-AIS Persian Gulf vessel population ranked by vessel count, each with a per-flag sanctioned tally, plus distinct-flag count and sanctioned share. Flags are derived best-effort from the MMSI MID (ITU-T M.585) and the payload is quality-gated on live AIS. Operators and inbound/outbound direction are omitted, not faked: no operator field is stored and per-vessel transit direction is not tracked.

The live-AIS Persian Gulf vessel population as one ranked, compliance-grade screen, worst-first. Every hull carries its risk band and the typed reasons behind it (ofac_sdn, ais_dark, foc_tanker, imo_spoofed: a tanker or cargo ship whose IMO fails its check digit) plus which sanctions bodies list it (US OFAC and/or UK/FCDO). A transparent rules engine, not ML, fusing OFAC SDN, live AIS, AIS-dark gaps, and detection of tankers and cargo ships whose IMO fails its check digit, which the flag rollup only aggregates and the single-vessel dossier resolves one IMO at a time. UK/FCDO listings show per hull and in counts.sanctioned; only an OFAC listing raises the band. counts cover the screened population, that is every hull seen on AIS inside the active window, not every hull ever tracked; the rows are the flagged hulls (high + moderate) by default, ?all=1 adds the cleared low band. Quality-gated on live AIS.

Current VLSFO, IFO380, and MGO delivered bunker prices ($/tonne) for Fujairah, benchmarked against Singapore and Rotterdam, plus the derived Fujairah premium over each hub. Fujairah is the world’s #2-3 bunkering hub, just outside the strait and the physical fueling alternative when transit is disrupted, so its spread over the global benchmarks is a physical-commodity read on Hormuz stress that the Brent and insurance signals do not capture. Current snapshot, checked every 6 hours; each port carries the date of its quote (priceDate), which is the last weekday print over a weekend.

A modeled war-risk insurance multiple (Nx over the peace baseline, on the published 1x to 80x scale) for moving a tanker through the strait. It’s recomputed every 10 minutes, but its inputs move at their own pace: throughput deviation comes from IMF PortWatch, which publishes weekly and can run more than a week behind; the escalation forecast (Polymarket, Kalshi, Manifold, Brent-WTI spread) and crude implied volatility (OVX) are live; event pressure comes from news feeds and GDELT over the last 72 hours. The insurance market’s own figure is deliberately kept out of the model, so the number is an independent read; the standing straits.live insurance-estimate multiple is carried alongside as a static reference (it is set manually and rarely changes), not as a live comparator. Also returns a modeled VLCC war-premium band and a confidence that degrades when driver feeds are stale or escalation-market volume is thin. A modeled situational-awareness indicator, not an insurance quote.