> ## Documentation Index
> Fetch the complete documentation index at: https://blackbox.dasha.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Validate Your Agent Before Deployment

> Test voice AI agents before deployment using browser testing, API calls, and debugging tools. Validate conversation quality, voice experience, and integrations.

Test systematically before going live. This guide covers testing workflow and checklists.

**What you'll learn:** Testing workflow, environments, and pre-launch checklists.

<Warning>
  Enabled agents receive live calls immediately. Test with the agent disabled before enabling.
</Warning>

***

## Testing workflow

### Phase 1: Configuration

<Steps>
  <Step title="Create disabled agent">
    Set `isEnabled: false` and configure LLM, voice, and system prompt
  </Step>

  <Step title="Preview voice">
    Compare voices, adjust speed (0.9-1.1x recommended)
  </Step>

  <Step title="Test prompts">
    Use chat mode to iterate on prompts quickly
  </Step>

  <Step title="Test webhooks">
    Verify endpoints return HTTP 200
  </Step>
</Steps>

### Phase 2: Conversation testing

<Steps>
  <Step title="Test happy paths">
    Run ideal conversation flows
  </Step>

  <Step title="Test edge cases">
    Try unclear input, interruptions, out-of-scope requests
  </Step>

  <Step title="Test errors">
    Disconnect webhooks, test timeout handling
  </Step>

  <Step title="Voice quality">
    Switch to voice mode, check clarity and latency
  </Step>
</Steps>

### Phase 3: Integration testing

<Steps>
  <Step title="Test tools">
    Trigger each tool, verify parameter extraction
  </Step>

  <Step title="Test MCP connections">
    Validate server connectivity
  </Step>

  <Step title="Test schedule">
    Verify business hours behavior
  </Step>

  <Step title="Test concurrency">
    Schedule multiple calls via API
  </Step>
</Steps>

### Phase 4: Pre-production

<Steps>
  <Step title="End-to-end voice test">
    Test all conversation branches
  </Step>

  <Step title="Review metrics">
    Target: \<2 seconds latency
  </Step>

  <Step title="Team review">
    Gather feedback on conversation quality
  </Step>

  <Step title="Enable for production">
    Monitor first 10-20 calls in Call Inspector
  </Step>
</Steps>

***

## Checklists

### Essential (every agent)

* [ ] Voice sounds natural and clear
* [ ] Core flows complete successfully
* [ ] System prompt instructions followed
* [ ] 10+ scenarios validated
* [ ] Error handling tested

### Advanced (complex agents)

* [ ] All tools execute successfully
* [ ] Webhooks deliver and retry
* [ ] MCP servers functional
* [ ] Multi-turn coherence maintained

### Edge cases (production-critical)

* [ ] Long silences handled (5+ seconds)
* [ ] Background noise tolerated
* [ ] Out-of-scope requests declined gracefully

***

## What's next

<CardGroup cols={2}>
  <Card title="Dashboard Testing" icon="browser" href="/docs/test/dashboard-testing">
    Test agents in the browser
  </Card>

  <Card title="Call Inspector" icon="magnifying-glass" href="/docs/test/call-inspector">
    Analyze completed calls in detail
  </Card>
</CardGroup>