opensre binary on developer machines, virtual machines, or on-prem
hosts.
What mode should I use?
Use the mode that matches where your workloads and alerts live:- Local laptop mode for fast setup, testing, and debugging
- On-prem VM/server mode for shared internal environments
- Container mode for reproducible deployments in Docker
Installation tab:
Local setup workflow
1) Install the OpenSRE CLI
sudo. It uses a writable bin directory already on your PATH when possible; otherwise it installs to ~/.local/bin and prints the command to apply the PATH update in your shell.
The installer makes opensre available on your PATH without sudo: it installs into (or links the binary from) a user-writable directory that is already on your PATH, so the command works in the current terminal right away. Only when no such directory exists does it fall back to updating your shell profile for new terminals.
When you run the installer from an interactive terminal — including the piped curl … | bash form above — it launches opensre onboard automatically once the install finishes, so you can set up your LLM provider and integrations right away. In non-interactive environments (CI, scripts, provisioning) the auto-launch is skipped and a “Next steps” hint is printed instead. To opt out of the auto-launch entirely:
2) Enter the OpenSRE shell
3) Run onboarding
From inside the OpenSRE shell:onboard helps you configure:
- LLM provider (for investigation reasoning)
- integration credentials
- optional communication/reporting integrations
4) Verify integration health
From inside the OpenSRE shell:5) Run your first investigation
From inside the OpenSRE shell (after running bareopensre):
Integration catalog
The full integration catalog is maintained under theIntegrations tab, not the Installation tab.
Start here:
Then open each provider page for credentials, minimum permissions, and setup examples.
Tracer