Documentation

Run AI agents like serverless functions. Describe the work, hit the API, get the result.

Quickstart

Get your API key and spin up your first agent container in under 5 minutes.

Core Concepts

Understand containers, agents, and tasks — the three building blocks of every workflow.

API Reference

Detailed documentation for the REST API endpoints, request bodies, and responses.

Billing & Credits

Prepaid credits for container time, BYOK for AI tokens, and what happens when your balance runs low.

Serverless agents

AgentContainer lets you run AI agents the way serverless lets you run functions — describe the work, hit the API, get the result. No containers to manage, no orchestration to build, no infrastructure to maintain. Each task spins up an isolated environment with Claude Code, the agent works autonomously, and the output is delivered back through the API.

You pay per task: container time at a flat per-minute rate. Bring your own AI provider key — token costs go straight to your provider, not through us. No idle costs, no subscriptions, no upfront commitment. EU-hosted infrastructure with containers running in Amsterdam.

How it works

The core model is three resources, each configured once and reused across tasks:

  1. Container — a reusable execution environment. Choose a standard preset or define your own.
  2. Agent — a reusable worker that runs inside a container. Set a prompt, a model, reference files, and sensible defaults.
  3. Task — a single unit of work dispatched to an agent. Describe what to do, attach input files if needed, and get results back.

Don't need custom configuration? Use a standard agent and dispatch tasks immediately — no setup required.

Dispatch a Taskbash
curl -X POST https://api.agentcontainer.com/api/v1/agents/agt_standard_coding/tasks \
  -H "Authorization: Bearer ac_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "description": "Read the attached PDF and extract all financial tables into a CSV."
  }'

What you get

Fully isolated containers

Every task runs in its own environment, spun up on demand and torn down on completion. No shared state between runs.

Autonomous agents

Claude Code works through your task end-to-end with structured completion logic. No babysitting, no polling a chat interface.

File in, files out

Upload input files before dispatch, download generated artifacts after. Whatever the agent produces is collected and stored for you.

Pay-per-task billing

Container time at a flat per-minute rate. Bring your own AI provider key — token costs are between you and your provider. No idle costs, no subscriptions.

Versioned configuration

Update an agent or container at any time. In-flight tasks continue on the version they were dispatched with. No surprises.

EU-hosted infrastructure

Containers run in Amsterdam. Your data stays in Europe.

Where to go from here

  • Quickstart — go from zero to a completed task in under 5 minutes
  • Core Concepts — understand how containers, agents, and tasks fit together
  • Tasks & Output — dispatching work, output policies, and retrieving results
  • API Reference — every endpoint, request body, and response schema

Ready to start building?

Create an account, load some credits, and dispatch your first task in minutes.

Follow the Quickstart