Travel advisory monitoring for tour stops

Canonical URL: https://signalguard.live/docs/signals/travel-advisory Meta description: SignalGuard reads US State Department travel advisories and attaches the right one to each tour stop. A US tour with a foreign date inherits that destination's Level 1–4 advisory; US and domestic stops resolve to "no advisory applies" and read as a clear signal. Built for tour security teams who plan across borders and need the federal travel posture per stop without checking travel.state.gov by hand.

What this signal monitors

SignalGuard is operated from the US perspective, and a tour routinely crosses borders — a US act with a London or Mexico City date needs the US State Department's travel advisory for that destination, not a generic one. The Travel advisory signal reads the State Department's published advisories and, for each stop on a tour, surfaces the advisory for that stop's country. The State Department issues no advisory for the United States itself, so a US or domestic stop resolves to "no advisory applies" — threatLevel is null, and the scan treats it as a clear signal rather than an error. The signal is per-stop, keyed by the venue, and it never guesses: an unmatched venue returns clear.

The four advisory levels

The State Department classifies every destination on a four-level ladder. SignalGuard parses the level directly from the advisory title:

• Level 1 — Exercise Normal Precautions. The lowest advisory. Standard care for any international trip.
• Level 2 — Exercise Increased Caution. Heightened risk; be aware of conditions.
• Level 3 — Reconsider Travel. Serious risk; reconsider whether to travel.
• Level 4 — Do Not Travel. The highest advisory. Life-threatening risk; the Department advises against all travel.

Data source

• Primary: the US State Department travel advisories feed (https://cadataapi.state.gov/api/TravelAdvisories), a public JSON list. Each item carries a Title (for example, United Kingdom - Level 2: Exercise Increased Caution), the advisory link, an HTML summary, and Published / Updated timestamps.
• Geocoding: a venue string is resolved server-side to its English country name via a geocoder, then matched against the feed. SignalGuard matches on the country name parsed from the advisory Title — not the State Department's category codes, which are GEC/FIPS codes that collide with ISO (the UK's code clashes with Gabon's). Matching on name avoids ever attaching the wrong country.
• Cache TTL: the advisory feed is cached for 6 hours and geocode results for 24 hours. State Department advisories are reissued periodically — event-driven, not real-time — so heavy caching keeps the signal fast without missing changes.

How SignalGuard scores severity

The level parsed from the advisory title maps straight onto SignalGuard's severity scale, which drives the chip color and the contribution to the venue threat score exactly like every other signal:

• Level 1 (Exercise Normal Precautions) → low
• Level 2 (Exercise Increased Caution) → medium
• Level 3 (Reconsider Travel) → high
• Level 4 (Do Not Travel) → critical

When a stop is in the US, is domestic, or can't be confidently matched to a country, threatLevel is null and the result is active: false — a clean, clear signal rather than an escalator. The signal only escalates a brief when there is an actual foreign-travel advisory to surface.

How it's wired

Travel advisory is a standard scan signal in the server pipeline, evaluated per stop and keyed by the venue. The endpoint is:

POST /api/travel-advisory
{ "location": "The O2, London" }     // venue string → geocoded to a country
{ "country":  "United Kingdom" }        // or a direct country-name lookup

A foreign match returns the canonical threatLevel (severity) plus the State Department level (1–4), its label (for example Reconsider Travel), the full advisory title, the advisory link, the resolved country, and the updatedAt timestamp, sourced from US State Department. A US, domestic, or unmatched stop returns ok: true with threatLevel: null and active: false. Upstream failure returns ok: false with code TRAVEL_ADVISORY_UPSTREAM_ERROR and is excluded from the rollup. The endpoint is rate-limited to 60 requests per minute.

Use cases for tour security

1. Foreign stop inherits the destination posture. A US tour plays a Level 3 destination mid-routing. That stop's brief escalates to high automatically, so the security director sizes the advance and on-site team to the country the act is actually in — not the country the tour is based in.
2. Domestic stops stay quiet. US dates resolve to "no advisory applies" and contribute nothing to the threat score. The Travel advisory chip is simply clear for those rows, and other live signals (weather, crime, social) carry the brief.
3. Routing review across borders. Reading the per-stop advisory across a whole tour at a glance shows which legs carry State Department risk — useful when deciding insurance, duty-of-care sign-off, and crew comms for international dates.

Coverage & limits

• Per-stop, not per-tour. The advisory is resolved at the stop level from the venue, so a multi-country tour can show different levels on different dates.
• Never guesses a country. If the geocoder can't resolve a venue confidently, or the country name doesn't match the feed, the signal returns clear rather than surfacing a best-guess country.
• No US advisory exists. The State Department does not issue an advisory for the United States, so domestic stops are clear by design — that is the correct result, not a gap.
• Reissue cadence. Advisories change when the State Department reissues them, which is periodic and event-driven rather than continuous. The cached feed reflects the latest published advisory.

Pairs well with

• News signal — local reporting behind a level change in a destination country
• NTAS signal — the domestic federal posture; when DHS escalates at home, the State Department often follows for outbound travel
• Disasters signal — natural-disaster context that can underlie a Level 3 or 4 advisory

FAQ

What does the Travel advisory signal do for a tour? For each stop, it surfaces the US State Department advisory for that stop's country. A foreign date inherits the destination's Level 1–4 advisory; US and domestic stops resolve to "no advisory applies" and read as clear.

How are the four levels mapped to severity? Level 1 (Exercise Normal Precautions) → low, Level 2 (Exercise Increased Caution) → medium, Level 3 (Reconsider Travel) → high, Level 4 (Do Not Travel) → critical. The mapping drives the chip color and severity like every other signal.

Does the signal ever attach the wrong country? No. The venue is geocoded to a country and matched by country name. On any uncertain match — or a US/domestic stop — the signal returns threatLevel: null and reads clear. SignalGuard never guesses a country.

How do I query it directly? POST /api/travel-advisory with { location } (a venue string, geocoded server-side) or { country } (a direct country-name lookup). The endpoint is rate-limited to 60 requests per minute.

How fresh is the advisory data? The State Department reissues advisories periodically — event-driven, not real-time. SignalGuard caches the feed for 6 hours, which reflects the latest published advisory while keeping the signal fast.