Skip to content
Clockwork
Notify me
for complex, custom builds

When one task is not enough.

The simple runner does one task. This is the full engine: many agents working together, with goals, budgets, and approvals. We build and run it for you.

Talk to me about a build
Works with
  • Claude Code
  • Codex

More on the way.

what you get

Everything a company needs. None of the headcount.

  • Bring your own agent

    Any agent, any runtime. If it can receive a heartbeat, it is hired.

  • Goal alignment

    Every task traces back to the mission. Agents know what to do, and why.

  • Heartbeats

    Agents wake on a schedule, check the work, and act. No tab to babysit.

  • Budgets

    A budget per agent. Hit the limit, they stop. No runaway spend.

  • Governance

    You are the board. Approve hires, override strategy, pause or stop any agent.

  • Org chart

    Roles, titles, reporting lines. Every agent has a boss and a job.

  • Full audit trail

    Every task threaded. Every decision logged. Nothing hidden.

  • Multi-company

    One deployment, many companies, fully isolated. Run a portfolio.

how it ships

Built and deployed for you.

We stand up the full engine, wire in your agents, and hand you the controls. It runs on your own infrastructure. You approve; it runs.

Boots in one command. Runs on standard infra.

clockwork
# server + embedded database, one command
$npx clockwork onboard --yes
✓ embedded postgres ready
✓ api at http://localhost:3100
→ hire your first agent
to be clear

What Clockwork is not.

  • Not a chatbotAgents have jobs, not chat windows.
  • Not a workflow builderNo drag-and-drop pipelines. Clockwork runs the operation.
  • Not a prompt managerAgents bring their own prompts and models.
  • Not for one taskThat is the simple runner. This is the whole operation.

Run the whole operation.

When you outgrow one task, we build the engine that runs the rest.

Talk to me about a build

Just need one task? See how the simple runner works →