Console
What each page of the log0 console does - incidents, the AI summary and investigation, and the settings pages for members, API keys, AI models, repositories and runbooks.
Incidents
The list filters and pages on the server (severity, status, environment, service name), so a workspace with thousands of incidents is fully reachable. Exact counts for the dashboard come from GET /api/v1/incidents/stats, counted in the database. The AI column shows what the investigation did about each incident.
Open an incident for its details. On a phone the panel is full screen.
The AI summary
Every new incident gets a short AI summary. It has three states, shown honestly:
| State | What you see |
|---|---|
| Ready | The summary text. |
| Generating | A skeleton, for at most three minutes after it was requested. |
| Unavailable | "No summary. The model did not answer." with Try again for admins and engineers (POST /api/v1/incidents/{id}/ai-summary/retry). |
The API returns this as aiSummaryStatus (READY, PENDING, UNAVAILABLE). A retry does nothing when a summary exists or one is already on its way.
Investigation
See AI investigations. The screen shows a progress rail, then tabs: Cause, Evidence, Fix and Steps, with feedback buttons, the run's model, time, tokens and cost, and a picker for earlier runs.
Settings (admins)
| Page | What it does |
|---|---|
| Workspace | Shows and renames the workspace (the slug never changes), and copies the tenant id and ingestion endpoint. |
| Members | Add a member, change a role (admin, engineer, viewer) or remove a member. You cannot change or remove yourself, and the last admin cannot be demoted or removed. A role change signs that person out of other devices so it takes effect. |
| API keys | Create and revoke ingestion keys; a key is shown once. |
| AI | Model, key, temperature and token limit per kind of work. |
| Repositories | Connect a GitHub repository, map services to folders, install the GitHub App and switch Draft PRs on per repository. |
| Runbooks | Add, read, download and delete runbooks. |
Logging out revokes your refresh token on the server, not only the browser cookie.
How is this guide?
Controlled Deploys
After you merge a log0 pull request, log0 can ask your own pipeline to deploy it, watch your logs to see whether the error went away, and roll back to the previous commit if you decide to. Every step needs a person, and nothing runs without being recorded.
Running Locally
Step-by-step guide to run the full log0 pipeline on your machine - the one-command Docker stack, environment variables, all seven services, health checks, log submission, and Redpanda/PostgreSQL inspection.