Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.sequentum.com/llms.txt

Use this file to discover all available pages before exploring further.

In this quickstart you’ll:
  1. Authenticate to the Sequentum API.
  2. Open an Agent Builder session and describe the data you want.
  3. Run the resulting agent and read records back in the same call.
You’ll need a Sequentum account and an API key. Keys are issued by an admin from Control Center → Users → Manage API keys. If you don’t have admin access, ask whoever does.

​
1. Set your API key

export SEQUENTUM_API_KEY="sk-..."

​
2. Start an Agent Builder session

Describe the agent you want in plain language. Agent Builder will plan, build, and test it for you. 
curl https://dashboard.sequentum.com/api/v1/agent-builder/start \
  -H "Authorization: ApiKey $SEQUENTUM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "prompt",
    "prompt": "Pull product name, price, and SKU from https://example-store.com/products/"
  }'
The response gives you a sessionId. The session is now running on Sequentum’s side.

​
3. Watch progress

Poll the events stream until the session reports completed. The endpoint accepts a lastEventIndex cursor — pass -1 on the first call and the highest index you’ve seen on each subsequent call. See Sessions for the full state machine. 
curl "https://dashboard.sequentum.com/api/v1/agent-builder/$SESSION_ID/events?lastEventIndex=-1" \
  -H "Authorization: ApiKey $SEQUENTUM_API_KEY"
When the session reports phase: "ready_to_finish", finalize it and a real Agent will be created. 
curl "https://dashboard.sequentum.com/api/v1/agent-builder/$SESSION_ID/finish" \
  -X POST \
  -H "Authorization: ApiKey $SEQUENTUM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "agentName": "Example store products" }'
The response includes the new agentId.

​
4. Run the agent and get records back inline

For the shortest path, run synchronously — the call blocks until the run completes and returns extracted records in the response body. 
curl "https://dashboard.sequentum.com/api/v1/agent/$AGENT_ID/start" \
  -X POST \
  -H "Authorization: ApiKey $SEQUENTUM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "isRunSynchronously": true }'
The response shape is { "<AgentName>": [ { ...record1 }, { ...record2 } ] }.
You just built and ran a web data agent end-to-end with three API calls.
For production, use async. Synchronous runs are subject to your HTTP client and our gateway timeouts. For anything that takes more than a few seconds, omit isRunSynchronously (or set it to false) — the call returns a runId immediately and you poll GET /run/{runId}/status until the status is terminal (Completed, Success, Failure, Stopped, or Skipped). See Runs for the full lifecycle and the Failure vs Failed distinction.

​
What’s next

Sessions in depth

The Agent Builder session state machine — phases, events, and the lifecycle from prompt to finished agent.

Schedule recurring runs

Attach a CRON, run-once, or interval schedule to any agent.

Production polling pattern

Skip isRunSynchronously and poll GET /run/{runId}/status until terminal — the pattern you want for anything beyond a few seconds.

Run lifecycle

Statuses, stop vs kill, what every run produces.