# Cekura

> Cekura is the testing and observability platform for voice AI agents. Run simulated conversations, evaluate performance with LLM-judge and code metrics, and monitor production calls.

## Getting Started
- [Introduction](https://docs.cekura.ai/documentation/introduction): Testing for AI Voice Agents. Launch in minutes not weeks by ensuring your agents deliver a seamless experience in every conversational scenario
- [Video Tutorials](https://docs.cekura.ai/documentation/video_tutorials): Explore our video tutorials to master AI voice agent testing and integration with Cekura.

## Key Concepts
- [Agent Overview](https://docs.cekura.ai/documentation/key-concepts/agents/overview): Introduction to the agents on Cekura
- [Setup Guide](https://docs.cekura.ai/documentation/key-concepts/agents/Agent_Setup_Guide): Step-by-step guide to creating and configuring an AI voice and chat agent on Cekura
- [Evaluator Overview](https://docs.cekura.ai/documentation/key-concepts/evaluators/overview): Introduction to Evaluators on Cekura
- [Conditional Actions](https://docs.cekura.ai/documentation/key-concepts/evaluators/conditional-actions): Create dynamic, rule-based evaluators that respond to specific conversation conditions
- [Personality](https://docs.cekura.ai/documentation/key-concepts/evaluators/personality): Understanding how personalities shape evaluator behavior in testing
- [Test Profile](https://docs.cekura.ai/documentation/key-concepts/evaluators/test-profile): Using test profiles to provide identity information for evaluators
- [Rubric Configuration](https://docs.cekura.ai/documentation/key-concepts/evaluators/rubric): Define custom success criteria for your evaluations
- [Mock Tools](https://docs.cekura.ai/documentation/key-concepts/evaluators/mock-tools): Learn how to create and manage mock tools for testing AI agents with predictable tool responses
- [Dynamic Variables](https://docs.cekura.ai/documentation/key-concepts/evaluators/dynamic-variables): Define agent-level variables that are automatically populated with scenario-specific values during workflow evaluator generation
- [Metrics Overview](https://docs.cekura.ai/documentation/key-concepts/metrics/overview): Metrics are used to understand and evaluate the performance of your AI Voice Agent
- [Pre-Defined Metrics](https://docs.cekura.ai/documentation/key-concepts/metrics/pre-defined-metrics): Standard metrics for evaluating voice agent performance across accuracy, conversation quality, customer experience, and speech quality
- [Output Types of Metric](https://docs.cekura.ai/documentation/key-concepts/metrics/output-types-of-metric): Understanding the different output types for metrics: boolean pass/fail, numeric scores, and categorical labels
- [LLM Judge Metric](https://docs.cekura.ai/documentation/key-concepts/metrics/llm-judge-metric): Evaluate AI voice agent calls using natural language criteria with LLM-powered judging
- [Available Variables](https://docs.cekura.ai/documentation/key-concepts/metrics/llm-judge-available-variables): Variables you can use in LLM Judge metrics
- [Python Metric](https://docs.cekura.ai/documentation/key-concepts/metrics/python-metric): Write custom evaluation logic in Python to evaluate your AI agent's performance with complete control
- [Metric Sampling](https://docs.cekura.ai/documentation/key-concepts/metrics/sampling): Reduce metric evaluation costs by evaluating only a percentage of calls
- [Twilio](https://docs.cekura.ai/documentation/key-concepts/phone-numbers/twilio): Bring your own Twilio phone numbers to Cekura for inbound and outbound testing. This guide walks you through importing and configuring Twilio numbers for seamless agent testing.
- [Plivo](https://docs.cekura.ai/documentation/key-concepts/phone-numbers/plivo): Bring your own Plivo phone numbers to Cekura for inbound and outbound testing.

## Integrations
- [Retell Testing](https://docs.cekura.ai/documentation/integrations/retell/testing): This guide will walk you through the process of setting up and testing voice agents using Retell integration in Cekura. When creating a new agent, you can select Retell as your provider from the avail
- [Retell Observability](https://docs.cekura.ai/documentation/integrations/retell/observability): Monitor and analyze your Retell-based voice agents
- [VAPI Testing](https://docs.cekura.ai/documentation/integrations/vapi/testing): This guide will walk you through the process of setting up and testing voice agents using VAPI integration in Cekura. When creating a new agent, you can select VAPI as your provider from the available
- [VAPI Observability](https://docs.cekura.ai/documentation/integrations/vapi/observability): Monitor and analyze your VAPI-based voice agents
- [ElevenLabs Testing](https://docs.cekura.ai/documentation/integrations/elevenlabs/testing): Test agents with automated ElevenLabs connection
- [ElevenLabs Observability](https://docs.cekura.ai/documentation/integrations/elevenlabs/observability): Monitor and analyze your ElevenLabs-based voice agents
- [LiveKit Automated](https://docs.cekura.ai/documentation/integrations/livekit/testing): Test agents with automated LiveKit room creation and token generation
- [LiveKit Tracing](https://docs.cekura.ai/documentation/integrations/livekit/tracing): Enhanced observability for LiveKit agents using the Cekura Python SDK
- [LiveKit Manual](https://docs.cekura.ai/documentation/integrations/livekit-integration): Run evaluator scenarios by manually providing LiveKit room URL and access token (voice and chat modes)
- [Pipecat Automated](https://docs.cekura.ai/documentation/integrations/pipecat/automated): Test Pipecat agents with automated session management
- [Pipecat Manual](https://docs.cekura.ai/documentation/integrations/pipecat/manual): Test Pipecat agents by providing Daily.co room URLs and tokens for each test run
- [SIP Calls](https://docs.cekura.ai/documentation/integrations/sip-integration): Testing SIP (Session Initiation Protocol) Based Agents
- [Custom Integration](https://docs.cekura.ai/documentation/integrations/custom-integration): Learn how to send call transcripts from your own backend system to Cekura using webhooks for simulation testing and observability.
- [Chat Testing](https://docs.cekura.ai/documentation/integrations/chat-testing): Test your voice agents faster and cheaper using text-based simulation

## Guides
- [Testing Overview](https://docs.cekura.ai/documentation/guides/testing-agents/overview): Learn how to quickly create and test voice agents on Cekura without requiring any external API keys or integrations.
- [Infrastructure Suite](https://docs.cekura.ai/documentation/guides/testing-agents/infrastructure-suite): Pre-built test scenarios to validate latency, stability, and failure handling across voice infrastructure stacks
- [Testing Approach](https://docs.cekura.ai/documentation/guides/testing-agents/suggested-testing-approach): A proven workflow to build reliable test cases for your AI voice agents
- [Outbound Calls](https://docs.cekura.ai/documentation/guides/testing-agents/outbound-evaluators): This guide walks you through the process of testing outbound calls using evaluators. You'll learn how to retrieve evaluators with phone numbers and execute outbound call tests.
- [Auto Outbound Calls](https://docs.cekura.ai/documentation/guides/testing-agents/outbound-auto-call): Automatically trigger outbound calls for testing without writing any scripts - the system handles everything using your voice provider configuration.
- [Load Testing](https://docs.cekura.ai/documentation/guides/testing-agents/load-testing): Learn how to perform load testing on your AI agents to ensure they can handle high call volumes
- [SMS During Call](https://docs.cekura.ai/documentation/guides/testing-agents/sms-during-call): Test SMS functionality with your voice agents - send and receive SMS messages during calls
- [WhatsApp Testing](https://docs.cekura.ai/documentation/guides/testing-agents/whatsapp-testing): Test conversational agents via WhatsApp messaging - fast and cost-effective text-based testing
- [DTMF Testing](https://docs.cekura.ai/documentation/guides/testing-agents/dtmf-testing): Test DTMF (keypad) interactions with your voice agents - simulate button presses during calls for IVR navigation and menu testing
- [Red Teaming](https://docs.cekura.ai/documentation/red-teaming/multi-turn): Test your AI agent against adversarial attacks with multi-turn conversation strategies

## Observability
- [Custom Observability](https://docs.cekura.ai/documentation/guides/observability/custom): To begin using observability features, you will need a Cekura account. If you do not have access, please contact [support@cekura.ai](mailto:support@cekura.ai).
- [Creating Good Metrics](https://docs.cekura.ai/documentation/guides/observability/creating-good-metric): Learn how to build and optimize custom metrics for AI agents through an iterative process of definition, testing, feedback, and optimization
- [Dynamic Description](https://docs.cekura.ai/documentation/guides/observability/dynamic-prompting): Learn how to use dynamic agent descriptions with Cekura's observability features to evaluate different agent configurations
- [PII Redaction](https://docs.cekura.ai/documentation/guides/observability/pii-redaction): Automatically detect and remove sensitive information from transcripts and audio recordings

## Automation
- [Dashboards](https://docs.cekura.ai/documentation/guides/dashboards): Build custom dashboards with widgets to visualize and analyze your call data, metrics, and metadata.
- [Organizing Your Workspace](https://docs.cekura.ai/documentation/guides/organizing-projects): Best practices for structuring projects, agents, and evaluators to maximize reusability and maintainability
- [GitHub Actions Tutorial](https://docs.cekura.ai/documentation/guides/github-actions-ci-cd): Learn how to set up GitHub Actions workflows to automatically test your agents with each change
- [Cron Job](https://docs.cekura.ai/documentation/guides/cronjob): Learn how to set up and manage automated cron jobs for your AI agent testing and evaluation workflows.
- [Prompting Examples](https://docs.cekura.ai/documentation/guides/prompting): This page provides examples of metrics and scenarios that can be used to test agent interactions. Each example includes both basic and advanced implementations, along with relevant scenarios and expec

## Agent Integration
- [Agent Integration](https://docs.cekura.ai/documentation/guides/agent-integration): How to use Cekura from AI coding agents like Claude Code, Cursor, and VS Code Copilot
- [AGENTS.md Template](https://docs.cekura.ai/documentation/guides/agents-md-template): A ready-to-use AGENTS.md template that teaches AI coding agents how to use Cekura in your project

## MCP Server
- [Model Context Protocol (MCP)](https://docs.cekura.ai/mcp/overview): Connect your AI assistant to Cekura for instant access to documentation and APIs
- [Skills](https://docs.cekura.ai/mcp/skills)

## Advanced
- [Enterprise Setup](https://docs.cekura.ai/documentation/advanced/enterprise-setup): Role-Based Access Control and API key management for enterprise deployments
- [IP Whitelisting](https://docs.cekura.ai/documentation/advanced/ip-whitelisting): Whitelist Cekura production IPs for firewalls, webhooks, and presigned URLs
- [Low Credits Alerts](https://docs.cekura.ai/documentation/advanced/credit-alerts): Get notified when your organization's credits fall below a threshold
- [Transcript Format](https://docs.cekura.ai/documentation/advanced/transcript-format): Example of transcript formats in Cekura
- [Webhook Format](https://docs.cekura.ai/documentation/advanced/webhook-format): Example of webhook formats in Cekura
- [Creating Custom Personalities](https://docs.cekura.ai/documentation/advanced/creating-custom-personalities): Personalities in Cekura define how your AI agents behave, communicate, and interact with users. They encompass the agent's tone, knowledge base, response style, and voice characteristics, allowing you
- [Knowledge Base Connectors](https://docs.cekura.ai/documentation/advanced/knowledge-base-connectors): Connect external data sources to your AI agents using knowledge base connectors. Automatically sync content from websites, databases, and other sources to keep your agent's knowledge up to date.
- [IVR & Voicemail](https://docs.cekura.ai/documentation/advanced/ivr-voicemail): Configure Interactive Voice Response (IVR) and voicemail scenarios using conditional actions and XML tags
- [Call End Reasons](https://docs.cekura.ai/documentation/advanced/call-end-reasons): Reference for all possible call end reasons in Cekura
- [Overview Embedding](https://docs.cekura.ai/documentation/embed/overview-embedding): Learn how to embed Cekura agent overview in your application using our embed widget. For token management details, see our [Token Generation](/embed/generate-token) and [Token Refresh](/embed/refreshi
- [Calls Embedding](https://docs.cekura.ai/documentation/embed/calls-embedding): Learn how to embed Cekura agent calls in your application using our embed widget. For token management details, see our [Token Generation](/embed/generate-token) and [Token Refresh](/embed/refreshing-
- [Result Embedding](https://docs.cekura.ai/documentation/embed/result-embedding): Learn how to embed Cekura agent results in your application using our embed widget. For token management details, see our [Token Generation](/embed/generate-token) and [Token Refresh](/embed/refreshin
- [Generate Token](https://docs.cekura.ai/documentation/embed/generate-token): Learn how to generate embed token for embedding your agent's result in your website
- [Refresh Token](https://docs.cekura.ai/documentation/embed/refreshing-expired-token): Learn how to refresh expired embed tokens for your agent's continuous operation
- [Changelog](https://docs.cekura.ai/documentation/advanced/changelog): Latest updates, new features, and improvements to the Cekura platform
- [FAQ](https://docs.cekura.ai/documentation/FAQ)

## API Reference
- [Calls](https://docs.cekura.ai/api-reference/observability/send-calls): Send, list, and evaluate production calls
- [Agents](https://docs.cekura.ai/api-reference/test_framework/get-agent): Create and manage agent configurations for testing
- [Mock Tools](https://docs.cekura.ai/api-reference/test_framework/list-mock-tools): Define simulated tool responses for evaluator scenarios
- [Metrics](https://docs.cekura.ai/api-reference/test_framework/get-metric): Create and manage LLM-judge and code-based evaluation metrics
- [Labs](https://docs.cekura.ai/api-reference/test_framework/post-test_frameworktest-setscreate_from_run): Metric improvement pipeline with feedback and auto-review
- [Evaluators](https://docs.cekura.ai/api-reference/test_framework/get-evaluator): Create test scenarios with personas, conditional actions, and expected outcomes
- [Test Profiles](https://docs.cekura.ai/api-reference/test_framework/get-test-profile): Reusable test configurations for evaluator runs
- [Results](https://docs.cekura.ai/api-reference/test_framework/get-result): View and manage evaluation run results
- [Runs](https://docs.cekura.ai/api-reference/test_framework/list-runs-with-ids): Inspect individual run details and bulk operations
- [Schedules](https://docs.cekura.ai/api-reference/schedules/get-schedule-job): Schedule recurring evaluation runs
- [Projects](https://docs.cekura.ai/api-reference/user/get-project): Organize agents, evaluators, and metrics into projects
- [Organization](https://docs.cekura.ai/api-reference/organization/get-billing-info): Billing and organization management
- [Dashboards](https://docs.cekura.ai/api-reference/dashboards/list-dashboards): Build custom dashboards for agent performance
- [Widgets](https://docs.cekura.ai/api-reference/dashboards/list-widgets): Configure dashboard widgets and data visualizations
- [API Keys](https://docs.cekura.ai/api-reference/user/create-api-key): Create and manage API keys for authentication
- [Others](https://docs.cekura.ai/api-reference/test_framework/list-personalities): Personalities, predefined metrics, phone numbers, and utilities

## Optional
- [OpenAPI Spec](https://docs.cekura.ai/openapi.json): Full OpenAPI 3.0 specification for all API endpoints