retellai-observability by jeremylongshore

---
name: retellai-observability
description: |
  Set up comprehensive observability for Retell AI integrations with metrics, traces, and alerts.
  Use when implementing monitoring for Retell AI operations, setting up dashboards,
  or configuring alerting for Retell AI integration health.
  Trigger with phrases like "retellai monitoring", "retellai metrics",
  "retellai observability", "monitor retellai", "retellai alerts", "retellai tracing".
allowed-tools: Read, Write, Edit
version: 1.0.0
license: MIT
author: Jeremy Longshore <jeremy@intentsolutions.io>
---

# Retell AI Observability

## Overview
Set up comprehensive observability for Retell AI integrations.

## Prerequisites
- Prometheus or compatible metrics backend
- OpenTelemetry SDK installed
- Grafana or similar dashboarding tool
- AlertManager configured

## Metrics Collection

### Key Metrics
| Metric | Type | Description |
|--------|------|-------------|
| `retellai_requests_total` | Counter | Total API requests |
| `retellai_request_duration_seconds` | Histogram | Request latency |
| `retellai_errors_total` | Counter | Error count by type |
| `retellai_rate_limit_remaining` | Gauge | Rate limit headroom |

### Prometheus Metrics

```typescript
import { Registry, Counter, Histogram, Gauge } from 'prom-client';

const registry = new Registry();

const requestCounter = new Counter({
  name: 'retellai_requests_total',
  help: 'Total Retell AI API requests',
  labelNames: ['method', 'status'],
  registers: [registry],
});

const requestDuration = new Histogram({
  name: 'retellai_request_duration_seconds',
  help: 'Retell AI request duration',
  labelNames: ['method'],
  buckets: [0.05, 0.1, 0.25, 0.5, 1, 2.5, 5],
  registers: [registry],
});

const errorCounter = new Counter({
  name: 'retellai_errors_total',
  help: 'Retell AI errors by type',
  labelNames: ['error_type'],
  registers: [registry],
});
```

### Instrumented Client

```typescript
async function instrumentedRequest<T>(
  method: string,
  operation: () => Promise<T>
): Promise<T> {
  const timer = requestDuration.startTimer({ method });

  try {
    const result = await operation();
    requestCounter.inc({ method, status: 'success' });
    return result;
  } catch (error: any) {
    requestCounter.inc({ method, status: 'error' });
    errorCounter.inc({ error_type: error.code || 'unknown' });
    throw error;
  } finally {
    timer();
  }
}
```

## Distributed Tracing

### OpenTelemetry Setup

```typescript
import { trace, SpanStatusCode } from '@opentelemetry/api';

const tracer = trace.getTracer('retellai-client');

async function tracedRetell AICall<T>(
  operationName: string,
  operation: () => Promise<T>
): Promise<T> {
  return tracer.startActiveSpan(`retellai.${operationName}`, async (span) => {
    try {
      const result = await operation();
      span.setStatus({ code: SpanStatusCode.OK });
      return result;
    } catch (error: any) {
      span.setStatus({ code: SpanStatusCode.ERROR, message: error.message });
      span.recordException(error);
      throw error;
    } finally {
      span.end();
    }
  });
}
```

## Logging Strategy

### Structured Logging

```typescript
import pino from 'pino';

const logger = pino({
  name: 'retellai',
  level: process.env.LOG_LEVEL || 'info',
});

function logRetell AIOperation(
  operation: string,
  data: Record<string, any>,
  duration: number
) {
  logger.info({
    service: 'retellai',
    operation,
    duration_ms: duration,
    ...data,
  });
}
```

## Alert Configuration

### Prometheus AlertManager Rules

```yaml
# retellai_alerts.yaml
groups:
  - name: retellai_alerts
    rules:
      - alert: Retell AIHighErrorRate
        expr: |
          rate(retellai_errors_total[5m]) /
          rate(retellai_requests_total[5m]) > 0.05
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "Retell AI error rate > 5%"

      - alert: Retell AIHighLatency
        expr: |
          histogram_quantile(0.95,
            rate(retellai_request_duration_seconds_bucket[5m])
          ) > 2
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "Retell AI P95 latency > 2s"

      - alert: Retell AIDown
        expr: up{job="retellai"} == 0
        for: 1m
        labels:
          severity: critical
        annotations:
          summary: "Retell AI integration is down"
```

## Dashboard

### Grafana Panel Queries

```json
{
  "panels": [
    {
      "title": "Retell AI Request Rate",
      "targets": [{
        "expr": "rate(retellai_requests_total[5m])"
      }]
    },
    {
      "title": "Retell AI Latency P50/P95/P99",
      "targets": [{
        "expr": "histogram_quantile(0.5, rate(retellai_request_duration_seconds_bucket[5m]))"
      }]
    }
  ]
}
```

## Instructions

### Step 1: Set Up Metrics Collection
Implement Prometheus counters, histograms, and gauges for key operations.

### Step 2: Add Distributed Tracing
Integrate OpenTelemetry for end-to-end request tracing.

### Step 3: Configure Structured Logging
Set up JSON logging with consistent field names.

### Step 4: Create Alert Rules
Define Prometheus alerting rules for error rates and latency.

## Output
- Metrics collection enabled
- Distributed tracing configured
- Structured logging implemented
- Alert rules deployed

## Error Handling
| Issue | Cause | Solution |
|-------|-------|----------|
| Missing metrics | No instrumentation | Wrap client calls |
| Trace gaps | Missing propagation | Check context headers |
| Alert storms | Wrong thresholds | Tune alert rules |
| High cardinality | Too many labels | Reduce label values |

## Examples

### Quick Metrics Endpoint
```typescript
app.get('/metrics', async (req, res) => {
  res.set('Content-Type', registry.contentType);
  res.send(await registry.metrics());
});
```

## Resources
- [Prometheus Best Practices](https://prometheus.io/docs/practices/naming/)
- [OpenTelemetry Documentation](https://opentelemetry.io/docs/)
- [Retell AI Observability Guide](https://docs.retellai.com/observability)

## Next Steps
For incident response, see `retellai-incident-runbook`.