HolmesGPT is a Python-based SRE agent (Apache-2.0) with operator mode for background incident investigation, integrating with multiple data sources and collaboration tools. It offers end-to-end automation for fetching alerts and writing findings back to sources like Slack, Jira, and GitHub.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
HolmesGPT — The CNCF SRE Agent. Open-source AI agent for investigating production incidents and finding root causes. Works with Kubernetes, VMs, cloud providers, databases, and SaaS platforms. Licensed under Apache-2.0.
How it works
Uses an agentic loop to query live observability data from multiple sources and identify root causes. Supports data sources including Prometheus, Kubernetes, Grafana, and many others. End-to-end automation can fetch alerts/tickets from external systems and write findings back to the source or Slack.
Getting started
Read the installation documentation linked in the repository homepage for CLI installation methods. The README mentions:
- Installation: link to installation documentation
- Using HolmesGPT: walkthroughs for Interactive mode and investigative workflows
Installation commands are not provided in the truncated README excerpt. Refer to the linked installation page for exact commands.
Recent releases
Latest release: 0.38.1 (2026-08-02) with changes including behavior-guard tests and fix permissions eval setup. Prior releases include 0.38.1-alpha (2026-07-29), 0.38.0 (2026-07-30), and 0.38.0-alpha (2026-07-29), and 0.37.0 (2026-07-27).
Traction
Stars: 2999. Forks: 428. Open issues: 374. Created: 2024-05-30. Last push: 2026-08-04.
Behind the repo
Not provided in the excerpt beyond general project background and affiliation with CNCF sandbox.
Caveats
License: Apache-2.0. Last push: 2026-08-04. Open issues: 374. Language: Python. Created: 2024-05-30.