Repolode
Submit
Tracer-Cloud

Tracer-Cloud/opensre

Open-source framework for building AI SRE agents that investigate production incidents across your own logs, metrics, traces and deploys.

View on GitHubWebsite
Stars
12,448
Forks
1,821
Open issues
66
Last commit
yesterday

What is it?

OpenSRE is an open-source framework for building AI site reliability engineering agents. When an alert fires or a engineer asks a question, the agent gathers correlated logs, metrics, traces, recent deploys and configuration from the tools already in use, then reasons through hypotheses in a tool-calling loop and returns an answer linked to the underlying evidence.

The project is available as an interactive terminal shell, a headless CLI for scripts and CI jobs, and a Python API for embedding the agent in existing services. Users can define their own runbooks, and the agent applies them during investigation. Optional reversible masking of identifiers such as pods, clusters and account IDs is available before calls to external LLMs.

OpenSRE connects to more than 60 tools, including Grafana, Datadog, Kubernetes, AWS, PagerDuty, Slack and several LLM providers. It is in public alpha, so APIs and integrations may still change.

Key features

  • 01Correlates logs, metrics, traces, deploys and config to diagnose incidents.
  • 02Links each root-cause conclusion to the underlying data.
  • 03Applies user-defined runbooks during investigation.
  • 04Offers an interactive shell, a headless CLI and a Python API.
  • 05Optionally masks identifiers before sending data to external LLMs.
  • 06Supports 60+ integrations across observability, cloud, databases and incident tools.

Use cases

  • Triaging production alerts by pulling evidence from existing Grafana, Datadog or Kubernetes setups.
  • Asking plain-language questions about why a service is slow during an incident.
  • Running scripted or CI-driven agent checks through the headless CLI.
  • Embedding incident investigation into internal Python automations and services.

Quick start

  1. 1

    Install opensre on macOS or Linux

  2. 2

    Start OpenSRE

    opensre

Things to know

  • Public alpha: core workflows are not yet fully stable and APIs may change.
  • Setup uses a curl install script rather than a package manager, and the first launch activates a hosted model.

In collections

Open-source AI agent frameworks & appsOpenTelemetry & observability platforms

Topics

#ai-agent#sre#incident-management#observability#root-cause-analysis#devops#llm#python#ai-sre#alerting

FAQ

What is opensre?
Open-source framework for building AI SRE agents that investigate production incidents across your own logs, metrics, traces and deploys. OpenSRE is an open-source framework for building AI site reliability engineering agents. When an alert fires or a engineer asks a question, the agent gathers correlated logs, metrics, traces, recent deploys and configuration from the tools already in use, then reasons through hypotheses in a tool-calling loop and returns an answer linked to the underlying evidence.
Is opensre free to use? What license does it have?
opensre is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is opensre still maintained?
Actively maintained. The last commit to opensre was on Oct 24, 2026.
What can I use opensre for?
Triaging production alerts by pulling evidence from existing Grafana, Datadog or Kubernetes setups. Asking plain-language questions about why a service is slow during an incident. Running scripted or CI-driven agent checks through the headless CLI. Embedding incident investigation into internal Python automations and services.
How do I install opensre?
Install opensre on macOS or Linux Start OpenSRE: opensre
Where can I get opensre?
The source code and installation instructions are on GitHub: https://github.com/Tracer-Cloud/opensre

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects