Queue status

Two lightweight read endpoints for operating at volume.

GET /browser-agent-status

Your own live queue: how many tasks you have running and queued, and your plan’s in-flight cap. Free to call, cheap on our side — poll it as often as you like.

curl https://api.pre.dev/browser-agent-status \
  -H "Authorization: Bearer $PREDEV_API_KEY"

{
  "userId": "...",
  "running": 12,
  "claimed": 3,
  "pending": 40,
  "total": 55,
  "cap": 100
}

Field	Description
`running`	Tasks executing right now
`claimed`	Tasks picked up, about to run
`pending`	Tasks waiting in your queue
`total`	running + claimed + pending
`cap`	Your plan’s max in-flight tasks — submits beyond this return `429 QUEUE_FULL`

Use it to throttle your own submit loop: stay under cap and you’ll never see QUEUE_FULL. Also available in the SDKs as browserAgentStatus() (TypeScript) and browser_agent_status() (Python).

GET /browser-agent-capacity

Global pool state — useful for deciding whether now is a good time to submit a large set of tasks. Any valid API key can read it.

curl https://api.pre.dev/browser-agent-capacity \
  -H "Authorization: Bearer $PREDEV_API_KEY"

Returns current queue and sandbox-pool statistics.

Run tasks

POST /browser-agent — sync, async, or streaming.

List runs

GET /list-browser-agents — history with pagination.

Stream a Running Task List Tasks

⌘I

Get Started

API Reference

SDKs

Migrations

Queue Status & Capacity

Queue status

GET /browser-agent-status

GET /browser-agent-capacity

Run tasks

List runs

​Queue status

​GET /browser-agent-status

​GET /browser-agent-capacity

Run tasks

List runs

Queue status

GET /browser-agent-status

GET /browser-agent-capacity