Agent Diagnostics
System health & execution trace
Live service checks, sandbox execution logs with retries, and the raw agent state graph.
Pipeline Duration
21.5s
Last full run
Total Tokens
17,062
Input + output
Avg LLM Latency
2583ms
Across 4 calls
Estimated Cost
$0.0312
This run
Service health
Backend API
Connected
42msjust now
Database
Connected
18msjust now
LLM Provider
Connected
612ms12s ago
Sandbox
Degraded
1840ms30s ago
Report Service
Connected
88ms1m ago
Generated code
plan = planner.invoke({
"profile": profile_summary,
"objective": "Explain revenue drivers and data quality risks",
})
for step in plan.steps:
print(step.id, step.description)stdout
1 Aggregate revenue by region 2 Trend revenue by month 3 Test discount depth vs units 4 Quantify missing data