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

# Introduction

> Build with the Bluejay simulations and observability APIs

Bluejay provides APIs for testing, monitoring, and improving conversational AI agents across voice and chat workflows.

## Base URL

```bash theme={null}
https://api.getbluejay.ai/v1
```

## Authentication

Sign up for a [Bluejay Account](https://app.getbluejay.ai) and
generate an API key from the [API Keys page](https://app.getbluejay.ai/settings/api-keys).
If you do not have a Bluejay account, book a demo [here](https://cal.com/rohanv/15min) to receive one.

Use that key to authenticate requests to the Bluejay API.

## Start here

<CardGroup cols={2}>
  <Card title="Run simulations" icon="flask" href="/api-reference/endpoint/queue-simulation-run">
    Queue realistic simulation runs to evaluate agents before they reach production.
  </Card>

  <Card title="Evaluate production calls" icon="chart-line" href="/api-reference/endpoint/evaluate">
    Measure quality, latency, and behavior in live conversations with observability workflows.
  </Card>

  <Card title="Manage agents" icon="robot" href="/api-reference/endpoint/add-agent">
    Create, update, and organize the agents that power your evaluation workflows.
  </Card>

  <Card title="Subscribe to webhooks" icon="bolt" href="/api-reference/webhook/events-webhook">
    Push results and event data into downstream systems when runs complete.
  </Card>
</CardGroup>

## Running a Simulation via API

<Steps>
  <Step title="Add Agent" titleSize="h3">
    [Add your agent](/api-reference/endpoint/add-agent) to Bluejay if you haven't already.
  </Step>

  <Step title="Create Simulation" titleSize="h3">
    [Create a simulation](/api-reference/endpoint/create-simulation) to group your Digital Humans so you can run them in parallel.
  </Step>

  <Step title="Generate or Create Digital Humans" titleSize="h3">
    [Generate](/api-reference/endpoint/generate-digital-humans) or [create](/api-reference/endpoint/create-digital-human) Digital Humans. Make sure to attach the Simulation ID so the Digital Humans are associated with the Simulation you created in Step 2.
  </Step>

  <Step title="Queue Simulation Run" titleSize="h3">
    [Queue a simulation run](/api-reference/endpoint/queue-simulation-run) to trigger the simulation — think of this as the Run button.
  </Step>
</Steps>

## Integrating Observability

<Steps>
  <Step title="Add Agent" titleSize="h3">
    [Add your agent](/api-reference/endpoint/add-agent) to Bluejay if you haven't already.
  </Step>

  <Step title="Define Custom Metrics" titleSize="h3">
    [Create custom metrics](/api-reference/endpoint/create-custom-metric) that capture the quality signals you care about — resolution rate, tone, compliance, or anything specific to your use case.
  </Step>

  <Step title="Send Calls to Evaluate" titleSize="h3">
    [Call the evaluate endpoint](/api-reference/endpoint/evaluate) with a recording URL or transcript from your production calls. Bluejay scores each call against your agent's configured metrics.
  </Step>

  <Step title="Review Results" titleSize="h3">
    [Retrieve call logs](/api-reference/endpoint/retrieve-call-logs) to inspect scores, latency data, and per-metric breakdowns across your evaluated conversations.
  </Step>
</Steps>

## Core API areas

* `Agents`: create and manage the systems you want to evaluate.
* `Simulations`: define test scenarios, digital humans, and simulation runs.
* `Observability`: evaluate live calls, logs, and quality signals.
* `Workflows`: automate evaluation and operational follow-up.
* `Webhooks`: stream Bluejay events into your own tooling.