Geographic Validation
Agent Status tests your agent from real devices around the world. Here's why that matters.
Why Geography Matters
Your agent might work perfectly in your office but fail for users in other regions due to:
- CDN edge issues — Content cached incorrectly in specific regions
- DNS propagation — Different DNS servers returning different IPs
- ISP routing — Some networks have poor routes to your servers
- Regional outages — Cloud provider issues in specific zones
- Latency — Users far from your servers experience slow responses
- Geo-blocking — Accidental IP blocking by region
Agent Status catches these issues before your users complain.
Available Regions
Agent Status's node network spans multiple continents:
| Region Code | Coverage |
|---|---|
us | United States, Canada |
eu | Western Europe, UK |
ap | Asia-Pacific (Japan, Singapore, Australia) |
latam | Latin America (Brazil, Mexico) |
af | Africa (Nigeria, South Africa, Kenya) |
me | Middle East |
Node availability varies by region. Check your dashboard for current coverage.
How It Works
When you configure regions for your agent:
Regions: ["us", "eu", "ap"]
Max Nodes: 10
Agent Status selects nodes matching your criteria:
- 3-4 nodes from US
- 3-4 nodes from EU
- 3-4 nodes from AP
Each node sends your prompts and evaluation prompts, measures latency, and reports back.
Per-Region Results
Your results include a geographic breakdown:
{
"verdict": "DEGRADED",
"by_region": {
"us": {"status": "UP", "latency_p50_ms": 245, "pass_rate": 100},
"eu": {"status": "UP", "latency_p50_ms": 312, "pass_rate": 100},
"ap": {"status": "DOWN", "latency_p50_ms": null, "pass_rate": 0}
}
}
This tells you: "Works in US and EU, failing in Asia-Pacific."
Tier Limits
Different subscription tiers have different node limits:
| Tier | Max Nodes per Run |
|---|---|
| Free | 3 |
| Hourly ($10/mo) | 10 |
| Standard ($20/mo) | 20 |
| Pro ($50/mo) | 50 |
More nodes = better coverage but more validations per check.
Use Cases
Global Production Monitoring
Regions: ["us", "eu", "ap", "latam", "af"]
Max Nodes: 30
Frequency: 15 min
Comprehensive coverage for global services.
US-Focused Service
Regions: ["us"]
Max Nodes: 5
Frequency: 1 hour
Lower cost, focused on primary market.
Pre-Launch Regional Testing
Regions: ["eu"]
Max Nodes: 10
Frequency: 5 min
Heavy monitoring during new region rollout.
Best Practices
Geo Context Injection
For agents with web search or location-aware features, enable inject_geo_context:
Inject Geo Context: ✅
This prepends the node's location to prompts:
[User location: Tokyo, Japan]
What's the weather like today?
Useful for testing if your agent gives different (correct) answers by location.
Related Articles
Agent vs Validator Errors
Not all failures are your fault. Agent Status distinguishes between problems with your agent and problems with our infrastructure.
Evaluation Prompts: Why They Matter
Evaluation prompts are Agent Status's secret weapon for detecting when an agent is 'up but broken.'
Latency & TTFB SLA
Fast responses matter. Agent Status tracks latency and enforces SLA thresholds to ensure your agent provides a good user experience.