Integration

TrainAR integration

TrainAR is an augmented reality training and field-service platform: engineers run guided sessions on AR glasses. With this API an agent can create, list and update operational tasks, list the account's users and invite new ones (an invitation email is sent), list, edit and delete recorded training sessions, run a skill against the account's knowledge base, manage form templates (own and library) and subscribe webhook endpoints to task, session and skill events. Users and sessions are personal data. The MCP server, OAuth app tokens and the partner (super-tenant) API are not exposed. Authentication is an account API key (prefix tak_) created by an admin in the TrainAR dashboard under API & Webhooks, API Keys, sent as a Bearer token; each key carries only the scopes ticked at creation.

Connect TrainAR How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in TrainAR.Built from the 18 TrainAR tools below, on the connection you add once.

What you can do with TrainAR

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Form, 6 operations

create_formCreate formPOSTAgents: ask firstCreates a form template for the account from a template image or PDF (by URL or base64), with optional field definitions. Needs the manage:forms scope.
delete_formDelete formDELETEAgents: ask firstDeletes one of the account's forms. Needs the manage:forms scope.
get_formGet formGETAgents: ask firstReturns one form of the account, or one form of the platform library, by id: name, aliases, category, active flag, template image URL and MIME type, field definitions. Needs the read:forms scope.
import_formImport formPOSTAgents: ask firstCopies a form of the platform library into the account's own forms and returns the new form. Find library forms with list_forms and library=true. Needs the manage:forms scope.
list_formsList formsGETAgents: ask firstLists the account's form templates (service sheets, safety records) that engineers fill in during sessions, or with library=true the platform form library. Needs the read:forms scope.
update_formUpdate formPUTAgents: ask firstUpdates one of the account's forms (name, aliases, active flag, category, fields, template URL); every field is optional. The docs do not say whether a field left out is kept, so read the form with get_form first. Needs manage:forms.

Session, 3 operations

delete_sessionDelete sessionDELETEAgents: ask firstDeletes a training session from the account (it leaves every list, report and tool at once) and propagates the deletion to the training app. Needs write:sessions, granted on request. Idempotent.
list_sessionsList sessionsGETAgents: ask firstLists the account's training sessions recorded on AR glasses (name, status, role, duration, start and end times, video URL, procedure count), page by page. Deleted sessions are left out. Personal data. Needs read:sessions.
update_sessionUpdate sessionPATCHAgents: ask firstEdits a recorded session's name, description, summary, duration, procedure count, video URL or thumbnail URL; only the fields sent change. Needs write:sessions, which TrainAR grants on request.

Task, 3 operations

create_taskCreate taskPOSTAgents: ask firstCreates an operational task in TrainAR (status open) and fires a task.created webhook to every subscribed endpoint. Needs the write:tasks scope.
list_tasksList tasksGETAgents: ask firstLists the account's operational tasks, newest first, with an optional filter on status, source, assigned user and creation date. Needs the read:tasks scope.
update_taskUpdate taskPATCHAgents: ask firstChanges a task's title, description, status, priority, due date or assignee; only the fields sent change. A status change fires task.status_changed (and task.completed when completed). Needs write:tasks.

Webhook, 3 operations

create_webhookCreate webhookPOSTAgents: ask firstSubscribes a public HTTPS URL to TrainAR events; TrainAR then POSTs each event there, signed with HMAC-SHA256. The signing secret is not returned: copy it from the endpoint's details in the TrainAR dashboard. Needs manage:webhooks.
delete_webhookDelete webhookDELETEAgents: ask firstUnsubscribes a webhook endpoint: TrainAR stops posting events to its URL. Fails with 404 when the endpoint is not found. Needs the manage:webhooks scope.
get_webhook_endpointGet webhook endpointGETAgents: ask firstChecks whether a webhook endpoint subscription still exists and returns its URL and events. Needs the manage:webhooks scope.

User, 2 operations

invite_userInvite userPOSTAgents: ask firstInvites a person to the TrainAR account: TrainAR sends an invitation email to the address at once, and the person sets a password on first login. Needs the manage:users scope.
list_usersList usersGETAgents: ask firstLists the account's users with email, name, role, status, assigned seat (glasses device) and session totals, page by page. Personal data. Needs the read:users scope.

Skill, 1 operation

execute_skillExecute skillPOSTAgents: ask firstRuns a TrainAR skill server to server against the account's knowledge base and returns the skill's answer (its shape depends on the skill). Fires a skill.executed webhook. Needs the write:skills scope.

Use TrainAR from Claude Code, Cursor or Codex

Connect TrainAR in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its TrainAR operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect TrainAR

Create your workspace
Sign up free. Nothing to install.
Connect TrainAR once
Add your TrainAR connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.