Total Calls (24h)
—
Total Cost (24h)
—
Error Rate
—
Avg Latency
—
48-Hour Trend
Agent Scorecards
| Agent | Score | Efficiency | Calls | Errors | Error % | Avg Latency | P95 | Cost | Trend |
|---|
Send Telemetry
Two ways to send agent data to Meridian:
OpenTelemetry (OTLP)
POST /api/telemetry/v1/traces
Content-Type: application/json
{ "resourceSpans": [{
"resource": {
"attributes": [{ "key": "service.name", "value": "my-agent" }]
},
"scopeSpans": [{
"spans": [{
"traceId": "abc123",
"name": "run_agent",
"durationNanoSeconds": 45000000,
"attributes": [
{"key":"cost_usd","value":0.0025},
{"key":"token_count","value":1250},
{"key":"tool.name","value":"web_search"}
]
}]
}]
}]}
Simple Format
POST /api/telemetry/events
Content-Type: application/json
{ "events": [{
"agent_name": "research-agent",
"trace_id": "abc123",
"event_type": "span",
"duration_ms": 3800,
"status": "success",
"cost_usd": 0.0032,
"token_count": 2100,
"tool_name": "web_search",
"tool_status": "success"
}]}
Agent Scorecard
Avg Latency (ms)
Cost (USD)
Calls
Cost Breakdown
Above cost threshold
7-day cost trend
Error Rate & Token Count
Error %
Tokens (k)
Recommendations
No recommendations for this agent.
Recent Alerts
No alerts for this agent.
Active Alerts
No active alerts. Your agents are healthy.
Registered Agents
No agents registered yet.
Cost Breakdown
Total Spend
—
Avg Cost / Call
—
Cost / 1K Tokens
—
Agents Tracked
—
Cost Over Time (per agent)
Spend by Agent
Per-Agent Cost Details
| Agent | Total Spend | % of Total | Calls | Cost / Call | Cost / 1K Tokens | Tokens |
|---|
Agent Leaderboard
Ranked Agents
| Agent | Score | Total Runs | Errors | Error % | Avg Cost / Run | Total Spend |
|---|
Run History
| Trace ID | Started | Duration | Agents | Outcome | Cost | Events | Timeline |
|---|
Run Comparison
Agent Divergence
| Agent | Cost A | Cost B | Cost Δ | Latency A | Latency B | Errors A | Errors B |
|---|
Interactive Demo
Load synthetic multi-agent trace data, then follow the three-step guided tour to explore key dashboard features: agent overview, cost breakdown, and reliability scoring.