Integration

Komos integration

Komos runs AI browser and workflow automations (tasks) that are built and trained in the Komos app. List and inspect the saved tasks of your organization with their input and output schemas, queue a run of a task with input values, follow its status, outputs and step-by-step logs, list runs across tasks, and cancel a run that is still pending or running. Building or editing tasks, schedules, credentials and webhooks is done in the Komos app, not through this public API. Authentication is an organization API key from the Komos dashboard, sent as a Bearer token.

Connect Komos How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Komos.Built from the 8 Komos tools below, on the connection you add once.

What you can do with Komos

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.

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

Run, 2 operations

list_task_runsList task runsGETReturns the 50 most recent runs of one task (from list_tasks) with status, inputs and outputs. Use list_runs to page or filter by status or date.
queue_task_runQueue task runPOSTQueues a run of a saved task and returns it at once with status PENDING; poll get_run with run.id until COMPLETED, FAILED or CANCELLED, then read run.outputs. Refused with HTTP 402 when your plan usage limit is reached.

Task, 2 operations

get_taskGet taskGETReturns one task by id (from list_tasks) with its declared inputs and outputs, schedule, node graph definition and AI plan summary.
list_tasksList tasksGETLists the saved tasks of your organization (id, name, status, declared inputs and outputs, schedule, run count), 20 per page by default, up to 100. Filter by name substring or status.

Task run, 2 operations

get_runGet runGETReturns the current state of one run: status (PENDING, RUNNING, COMPLETED, FAILED, CANCELLED), progress, inputs, outputs once completed, and error when it failed.
list_runsList runsGETLists runs across all tasks of your organization, newest first, 20 per page by default, up to 100. Filter by status, taskId and a created-at window (since defaults to 7 days ago).

Cancel, 1 operation

cancel_runCancel runPOSTRequests cancellation of a run that is still PENDING or RUNNING and returns the run; a run already in a final status comes back unchanged. Poll get_run for the final status.

Log, 1 operation

get_run_logsGet run logsGETReturns the ordered log entries a run captured (level, message, node, and per-step inputs and outputs when available).

Use Komos from Claude Code, Cursor or Codex

Connect Komos in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Komos 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 Komos

Create your workspace
Sign up free. Nothing to install.
Connect Komos once
Add your Komos 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.