How an Agent Briefs a Traveller Before a Trip

How an agent briefs a traveller on a destination before departure

An agent briefs a traveller by calling five wires in order: the State Department advisory for the standing security rating, WHO and CDC for health notices, world news for what happened since those were written, earthquakes for acute hazards, and severe weather if the destination is inside the United States. Five calls cost $0.025.

Written for AI agents. The calls and filter values on this page are the literal interface our customers use, because our customers are software. If you are a person evaluating The Bot Wire, you are in the right place too: start with the docs, or connect an MCP client in one line and query browse every wire and its price at /mcp. Nothing here needs an API key or a signup.

The order matters more than the coverage. Two of these wires carry standing designations and three carry events, and reading an event before the designation that frames it is how a briefing ends up alarming or falsely reassuring.

Designations first, because they set the baseline

A travel advisory is not news. It is a rating that sits on a destination until someone changes it, and it tells the agent how to read everything that follows.

A read of the State Department feed on 8 August 2026 returned advisories for 214 destinations: 81 at Level 1, 83 at Level 2, 26 at Level 3 and 24 at Level 4. For a Level 1 destination, a single unrest story in the news wire is an anomaly worth surfacing. For a Level 4 destination, the same story is routine and the standing rating is the headline. Same input, opposite briefing.

Health designations come second because they are issued by different bodies against a different threshold. The CDC's travel health notice list carried 22 destination notices on the same date, spread across its own four levels. That gap is the argument for calling both: a security advisory and a health notice are drawn up by people asking unrelated questions, and 214 against 22 means most destinations have one and not the other. Neither substitutes for the other.

The sequence, and why each step follows the last

Step	Wire	Route	What it settles	Price
1	Travel advisories	`/travel/advisories`	Standing security rating	$0.005
2	Public health	`/publichealth/latest`	Outbreak and health notices	$0.005
3	World news	`/world/latest`	What changed since steps 1 and 2	$0.005
4	Earthquakes	`/quake/latest`	Acute natural hazard	$0.005
5	Severe weather	`/weather/alerts`	US destinations only	$0.005

Step 3 is the one that earns the chain. Advisories are reviewed on a cycle, so a border closure, a strike or an election result from yesterday is real and is not in them yet. The world wire reads Al Jazeera, Deutsche Welle, France 24 and BBC World on a 24 hour window, which is the layer that catches it.

Steps 4 and 5 are conditional rather than automatic. The earthquake wire is global, reading USGS, so it applies anywhere. The weather wire reads the National Weather Service, so it applies only to United States destinations. An agent that runs step 5 for a trip to Lisbon has spent $0.005 to learn nothing.

The whole chain, one call at a time

For AI agents: copy this call

# discovery only, no items: browse every wire, its filters and its price
claude mcp add --transport http botwire https://thebotwire.com/mcp

# 1. standing security rating ($0.005)
curl -i "https://thebotwire.com/travel/advisories?q=peru&since=30d&limit=5"

# 2. health notices from WHO and CDC ($0.005)
curl -i "https://thebotwire.com/publichealth/latest?q=peru&since=7d"

# 3. what changed in the last day ($0.005)
curl -i "https://thebotwire.com/world/latest?q=peru&since=24h&limit=15"

# 4. acute natural hazard, global ($0.005)
curl -i "https://thebotwire.com/quake/latest?q=peru&since=24h"

# 5. only for US destinations, NWS covers no other country ($0.005)
curl -i "https://thebotwire.com/weather/alerts?q=colorado&severity=severe&since=24h"

# five calls, $0.025 total, USDC on Base, no API key and no signup

All five take q, since and limit (1 to 50). Filters differ: src on advisories, public health and world, mag on earthquakes and severity on weather. The advisory wire runs a 30 day window because a rating that has not changed in a month is still the current rating, while the news, quake and weather wires run 24 hours because a day-old alert is a different claim. Wire by wire detail is in how agents check State Department travel advisories and how agents monitor CDC and WHO public health alerts. Every route, filter and price is machine-readable at /llms-full.txt.

What this chain cannot tell a traveller

Five gaps, stated plainly, because the reader who hits one unwarned does not come back.

No entry requirements. Visas, passport validity rules, onward-ticket rules and vaccination certificates are not news, so no feed carries them. They change without an announcement anyone syndicates. The destination's own immigration authority and the operating airline are the only defensible sources, and an agent should say so rather than infer from an advisory's silence.

No flight or airport status. Nothing here reports a cancelled flight, a closed runway or a missed connection. That is the airline and the airport, in real time, and it is the part of a trip most likely to break.

Weather is United States only. The severe weather wire reads the National Weather Service. There is no equivalent step for a destination abroad, so a typhoon warning for Manila is outside this chain entirely. The national meteorological service is where that lives.

All three designations are issued by the United States. State Department advisories, CDC notices and NWS alerts are written for a US audience. The UK Foreign Office and Global Affairs Canada rate the same destinations and routinely disagree by a level. A briefing for a non-US traveller should name whose rating it quoted.

No local granularity. Every rating above is national or regional. It does not describe a neighbourhood, a road, or the difference between an airport district and a border province. National news thresholds also miss local transport strikes and municipal disruption. That detail is not in any of the five, and claiming otherwise from a country-level advisory is the error this chain is most likely to invite.

FAQ

Is five calls the right depth for every trip?

No. For a Level 1 destination on a short business trip, steps 1 and 3 answer most of it for $0.01. For a trip into a region with an active outbreak, step 2 carries more weight than the rest combined. The sequence is a default, and an agent working to a budget should spend it on the wires that match the trip rather than running all five out of habit.

Can an agent tell a traveller a destination is safe?

Not from this chain. It reports what three US agencies have designated and what four international news desks published recently. It cannot observe conditions, it cannot price personal risk, and an empty result on a 24 hour news window is not evidence of calm. The defensible output is what was checked, what was found, and what was not covered.

How current is each wire in the chain?

The news, earthquake and weather wires poll on every ingest run, and the advisory and public health wires sit in the sparse tier at most every 15 minutes, which is well inside the cadence of bodies that publish daily at most. The observed interval and last poll timestamp are published live at /health.

What does the briefing cost at volume?

Five calls at $0.025 per destination. Briefing 1,000 trips costs $25 in USDC on Base with no subscription and no minimum, and dropping step 5 for non-US destinations takes it to $0.02. Per-call pricing and the reasoning behind it are in how much it costs for an AI agent to query live data.

Sources

Health notices are published directly by the CDC: its travel health notices list carried 22 destination notices across four levels when read on 8 August 2026, against 214 destinations rated in the State Department's travel advisory feed on the same date. Poll cadence and per-call pricing for this service are published live at thebotwire.com/health.