MCP Server

Bluejay exposes a Model Context Protocol (MCP) server that gives AI assistants — Claude, Cursor, Windsurf, and any other MCP-compatible client — direct, structured access to the full Bluejay API. Instead of writing HTTP calls by hand, your AI assistant can run simulations, manage agents, queue test runs, and retrieve results as native tool calls.

MCP Endpoint

https://api.getbluejay.ai/mcp

The server uses stateless HTTP (MCP Streamable HTTP transport). Every request to /mcp is a standard HTTP POST that returns the tool result inline. No persistent connection or SSE stream is required.

Authentication

All requests to the MCP server must be authenticated. The recommended method is an API key passed in the X-API-Key header.

Get your API key

Sign in to app.getbluejay.ai
Navigate to Settings > API Keys
Click Generate API Key and copy the value immediately — it is only shown once

Pass your API key

Include your key in the X-API-Key header on every request to the MCP server:

X-API-Key: <your_api_key_here>

Error responses

Status	Meaning
`401`	Missing or invalid credential
`403`	Valid JWT but user has no organization

Unauthenticated requests receive a JSON body: { "error": "Unauthorized", "detail": "<reason>" }.

Configuring your MCP client

Claude Code & Claude Desktop

Set up Bluejay MCP in Claude Desktop or Claude Code CLI

Cursor & Windsurf

Set up Bluejay MCP in Cursor or Windsurf

Available tools

The MCP server injects your identity from the X-API-Key header automatically — you never pass credentials as tool arguments. The families below cover the most common tools; the server exposes more (the full list comes back from your MCP client’s tool discovery).

Agents

Tool	Description
`add_agent`	Create a new agent
`update_agent`	Update an agent by ID
`update_agent_by_external_id`	Update an agent by your own external ID
`get_agent`	Fetch a single agent
`get_agent_by_external_id`	Fetch an agent by external ID
`get_agents_by_folder`	List agents in a folder
`list_agents`	List all agents in your organization
`delete_agent`	Delete an agent

Simulations

Tool	Description
`create_simulation`	Create a new simulation
`update_simulation`	Update a simulation
`get_simulation`	Fetch a simulation
`list_simulations`	List all simulations
`get_simulations_by_agent`	List simulations for a given agent
`get_simulation_runs`	List runs for a simulation
`delete_simulation`	Delete a simulation

Digital Humans

Tool	Description
`create_digital_human`	Create a single Digital Human
`bulk_create_digital_humans`	Create multiple Digital Humans at once
`generate_digital_humans`	AI-generate Digital Humans for a simulation
`update_digital_human`	Update a Digital Human
`get_digital_human`	Fetch a Digital Human
`get_digital_humans_by_simulation`	List Digital Humans for a simulation
`delete_digital_human`	Delete a Digital Human

Simulation Runs

Tool	Description
`queue_simulation_run`	Queue a voice simulation run
`queue_sms_simulation_run`	Queue an SMS simulation run
`queue_http_text_simulation_run`	Queue an HTTP/text simulation run
`queue_email_simulation_run`	Queue an email simulation run
`get_simulation_results`	List results for a simulation run
`get_simulation_result`	Fetch a single simulation result

Evaluation & Observability

Tool	Description
`evaluate`	Submit a production call for evaluation
`send_http_text_message`	Send a message in an active HTTP/text simulation
`translate_transcript`	Translate a call transcript
`end_conversations`	End one or more active conversations
`delete_call_log`	Delete a call log
`update_call_log`	Update a call log

Custom Metrics

Tool	Description
`create_custom_metric`	Create a custom evaluation metric
`create_custom_metrics`	Bulk-create custom metrics
`update_custom_metric`	Update a custom metric
`get_custom_metric`	Fetch a custom metric
`list_custom_metrics`	List all custom metrics
`get_custom_metrics_by_agent`	List custom metrics for an agent
`delete_custom_metric`	Delete a custom metric
`bulk_delete_custom_metrics`	Delete multiple custom metrics

Folders

Tool	Description
`create_folder`	Create a folder
`update_folder`	Update a folder
`get_folder`	Fetch a folder
`list_folders`	List all folders
`move_agent_to_folder`	Move an agent into a folder
`delete_folder`	Delete a folder

Agent Versions

Tool	Description
`create_prompt_version`	Create a new agent version
`get_prompt`	Fetch an agent version
`list_prompts`	List all agent versions
`list_prompt_versions`	List an agent’s versions
`get_prompt_version`	Fetch a specific agent version
`add_prompt_version_labels`	Add labels to an agent version
`get_prompt_version_labels`	Get labels for an agent version
`delete_prompt_version_labels`	Remove labels from an agent version
`delete_prompt`	Delete an agent version by unique ID
`delete_prompt_version`	Delete an agent version by version number

Labels

Tool	Description
`create_label`	Create a label
`list_labels`	List all labels
`get_label`	Fetch a label
`update_label`	Update a label
`delete_label`	Delete a label

Communities

Tool	Description
`create_community`	Create a community
`update_community`	Update a community
`get_community`	Fetch a community
`list_communities`	List all communities
`delete_community`	Delete a community

Workflows

Tool	Description
`create_workflow`	Create a workflow
`update_workflow`	Update a workflow
`get_workflow`	Fetch a workflow
`list_workflows`	List all workflows
`delete_workflow`	Delete a workflow

Schedules

Tool	Description
`create_schedule`	Schedule a recurring simulation run
`update_schedule`	Update a schedule
`get_schedule`	Fetch a schedule
`list_schedules`	List all schedules
`delete_schedule`	Delete a schedule

Phone Numbers

Tool	Description
`list_phone_numbers`	List phone numbers in your organization
`add_phone_number`	Add a phone number
`release_phone_number`	Release a phone number

Alerts

Tool	Description
`create_alert`	Create an alert on a metric
`update_alert`	Update an alert
`list_alerts`	List alerts
`get_alert_results`	Fetch alert firing history
`delete_alert`	Delete an alert

Uptime Monitors

Tool	Description
`create_uptime_monitor`	Create a monitor (provisions its health-check simulation and schedule)
`update_uptime_monitor`	Partial update — checks, notifications, frequency, pause/resume
`list_uptime_monitors`	List monitors with checks, notifications, and last status
`get_uptime_monitor`	Fetch one monitor and its recent checks
`delete_uptime_monitor`	Delete a monitor and stop its checks

Checks and notifications use the same shapes as the REST API, including the digital-human custom caller.

Agent Workflows (provider sync)

Tool	Description
`sync_agent_workflow`	Pull the latest workflow from the provider
`push_agent_workflow`	Push edited workflow back to the provider
`get_workflow_summary`	Fetch the synced workflow graph
`list_workflow_paths`	Enumerate distinct paths through the workflow
`add_workflow_node` / `update_workflow_node` / `delete_workflow_node`	Edit nodes
`add_workflow_edge` / `delete_workflow_edge`	Edit edges

Self-Improve

Tool	Description
`start_self_improve_run`	Start an autonomous prompt-improvement loop
`get_self_improve_run`	Fetch run status and per-round history
`list_self_improve_runs`	List runs
`stop_self_improve_run`	Stop a run

Other families

The server also exposes tools for Metrics Lab (labs, annotations), Traces & Spans, Scenario nodes (IVR, hold, human-agent), Voice options, and provider-specific helpers for Vapi (squads, assistant tools) and ElevenLabs (agents, branches, merge).

Example: Run a simulation end-to-end

With the MCP server connected, you can ask your AI assistant:

“List my agents, pick the one named ‘Support Bot’, create a simulation called ‘Regression Test’, generate 5 Digital Humans, and queue a simulation run.”

The assistant will chain the list_agents → create_simulation → generate_digital_humans → queue_simulation_run tools automatically, without you writing any code.

Overview

Guides

MCP Endpoint

Authentication

Get your API key

Pass your API key

Error responses

Configuring your MCP client

Claude Code & Claude Desktop

Cursor & Windsurf

Available tools

Agents

Simulations

Digital Humans

Simulation Runs

Evaluation & Observability

Custom Metrics

Folders

Agent Versions

Labels

Communities

Workflows

Schedules

Phone Numbers

Alerts

Uptime Monitors

Agent Workflows (provider sync)

Self-Improve

Other families

Example: Run a simulation end-to-end

​MCP Endpoint

​Authentication

​Get your API key

​Pass your API key

​Error responses

​Configuring your MCP client

Claude Code & Claude Desktop

Cursor & Windsurf

​Available tools

​Agents

​Simulations

​Digital Humans

​Simulation Runs

​Evaluation & Observability

​Custom Metrics

​Folders

​Agent Versions

​Labels

​Communities

​Workflows

​Schedules

​Phone Numbers

​Alerts

​Uptime Monitors

​Agent Workflows (provider sync)

​Self-Improve

​Other families

​Example: Run a simulation end-to-end

MCP Endpoint

Authentication

Get your API key

Pass your API key

Error responses

Configuring your MCP client

Available tools

Agents

Simulations

Digital Humans

Simulation Runs

Evaluation & Observability

Custom Metrics

Folders

Agent Versions

Labels

Communities

Workflows

Schedules

Phone Numbers

Alerts

Uptime Monitors

Agent Workflows (provider sync)

Self-Improve

Other families

Example: Run a simulation end-to-end