AGENT SKILL

retellai-observability

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".

License 许可
MIT
最近更新
2026-11-15
Stars
★ 2.0k
Forks
233

01 前言 Front Matter

name
retellai-observability
author
Jeremy Longshore <jeremy@intentsolutions.io>
license
MIT
version
1.0.0
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

03 SKILL.md 正文

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

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

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

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

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

# 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

{
  "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

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

Resources

Next Steps

For incident response, see retellai-incident-runbook.

04 同仓库更多 Skills

Dicklesworthstone
Dicklesworthstone/pi_agent_rust

theme-factory

Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.

视觉设计
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

web-artifacts-builder

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.

前端开发
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

测试与质量前端开发
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

neuropixels-analysis

Neuropixels neural recording analysis. Load SpikeGLX/OpenEphys data, preprocess, motion correction, Kilosort4 spike sorting, quality metrics, Allen/IBL curation, AI-assisted visual analysis, for Neuropixels 1.0/2.0 extracellular electrophysiology. Use when working with neural recordings, spike sorting, extracellular electrophysiology, or when the user mentions Neuropixels, SpikeGLX, Open Ephys, Kilosort, quality metrics, or unit curation.

★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

skill-creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Skill与智能体开发
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

notebooklm

Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.

Skill与智能体开发
★ 2.0k