01
See what needs attention.
Start with project health, recurring failures and missing workers. Move from an alert to the task history behind it.
Investigate failures Open source · Self-hosted · v1.11.0 Background work.
Background work.
Clearly in view.
One control plane for your Python task queues. Find the failure, understand the schedule, and take the next step with context.
The demo opens with sample projects. No account required.
From signal to action
Less searching.
More understanding.
02
Keep scheduled work on track.
Understand when a task should run, what happened last time and how missed runs recover. Manage schedules alongside the work they create.
Explore schedules03
Act with context.
Use capability-aware commands and governed automation. Follow command delivery and inspect the audit history after an action.
Explore automationOne connected workspace
The task is only
the beginning.
Keep task history, workers, queues, schedules and automation together. Switch projects without losing your place, and follow an investigation from a recurring issue to the underlying execution.
Explore the platform →
Your infrastructure
Run it where your work runs.
Start with pip and SQLite, use a container, or deploy with PostgreSQL. The brain runs in your environment. Agent packages use Apache 2.0; the server and dashboard use AGPL-3.0-or-later.
Operational controls, visible boundaries.
Role-based access, MFA, redaction and tamper-evident audit history are part of the product. z4j publishes its threat model and known limitations. Read the deployment guidance before running production workloads.
See your next step.
Explore an incident in the demo, then connect your own worker.