Investigate production incidents across Kubernetes and cloud signals with HolmesGPT
Use HolmesGPT when an on-call agent needs one investigation loop that pulls alerts, logs, metrics, and infrastructure context from multiple systems and returns a root-cause path instead of forcing a human to hop across separate observability products.
Prerequisites
HolmesGPT CLI or operator deployment, one supported LLM provider, and connected observability/toolset integrations
Installation
Requirements and caveats from upstream:
Basic usage or getting-started notes:
Extracted from upstream docs: https://raw.githubusercontent.com/HolmesGPT/holmesgpt/HEAD/README.md