Developers

Everything the console does, over HTTP.

Create and update agents, start runs, read run timelines, manage tools, decide approvals and subscribe to events. The console is a client of this API rather than a privileged surface, so there is nothing it can do that you cannot.

Read the docs Quickstart

The console has no private endpoints

If the interface can do it, the API can. A platform whose own console uses capabilities its customers cannot reach will always drift towards leaving developers behind.

Deploying still goes through review

The API can create and update an agent. Bringing one into production still produces the review record and requires the approval, because a path around the review screen is a path around the security model.

Runs are readable, not just startable

The full run timeline, every tool call, every decision, every approval and its cost. The same record the console shows, in the same order.

Scoped keys, per environment

API credentials are scoped to an environment and to a set of capabilities. A staging key cannot reach production, and a key that can read runs cannot deploy an agent.

Roughly what it looks like

Illustrative and deliberately short. The reference is at docs.agentlabora.com, and pretending a marketing page is documentation wastes an afternoon.

Start a run and read what happened

http

POST /v1/agents/invoice-summary/runs
Authorization: Bearer $LABORA_KEY

{ "input": { "invoice": "INV-4421" } }

GET /v1/runs/run_8f21/timeline

Where it stops

The API cannot grant an agent a capability that is refused at the platform, cannot approve its own requests and cannot read another organisation's data — those are not permission questions, they are refusals. Approvals in particular cannot be granted by an API key: the point of an approval is that a person decided.

Developer platform

The rest of it.

QuickstartA deployed agent in under fifteen minutes.SDKsPython and TypeScript, maintained in step.CLICreate, test, deploy, tail logs, roll back.Custom toolsExpose your own service as a permissioned capability.Agent manifestThe declarative file your agent lives in.SandboxSynthetic data, mock tools, no production reach.EvaluationsTest suites that run in your pipeline.PublishingShip an agent to the marketplace.DocumentationThe reference, the guides and the API explorer.

Give AI a job.

Build an agent, choose what it can access, define what it may do, and deploy it to work safely across your business.

Build an agent Build with the API

No code required · Full SDKs and API · Stop any agent instantly · [email protected]