holmesgpt

Discover HolmesGPT safeguards for safe tool execution. Learn about its defense-in-depth architecture including validation, filtering, approval workflows, and endpoint restrictions.

Add a custom LLM model to HolmesGPT's model registry by configuring YAML. Integrate any LiteLLM supported endpoint seamlessly without code changes.

Learn to configure PagerDuty or OpsGenie integration in HolmesGPT using API keys and config files. Streamline your incident management with automatic plugin instantiation.

Master HolmesGPT interactive mode for seamless infrastructure troubleshooting. Use the `holmes ask` REPL to chat with AI, execute commands, and share outputs directly from your terminal.

Explore the core architectural difference in HolmesGPT. Understand how sources ingest alerts and toolsets provide executable commands for LLM investigations.

Learn to configure Slack output in HolmesGPT using CLI flags environment variables or a YAML file. Integrate your workflows seamlessly with this complete setup guide.

Explore HolmesGPT's custom toolsets vs CLI experimental toolsets. Learn to extend capabilities effectively through stable overrides or supplementary CLI options.

Discover how the HolmesGPT prompt building system works. Learn about its two-stage pipeline architecture for configurable system and dynamic user prompts using Jinja2 and runtime enrichment.

Learn to add a new data source toolset to HolmesGPT by creating a YAML definition, registering it, and refreshing the tool cache for seamless integration.

Learn when to use the JSON filter mixin in HolmesGPT custom toolsets. Effectively manage large or nested JSON payloads to avoid token limits and streamline your investigations.

Master HolmesGPT max_steps configuration for investigations. Learn to set limits via CLI, config files, and code examples for efficient tool-calling.

Discover how HolmesGPT manages investigation state in memory using a single loop within the ToolCallingLLM class. Learn about its architecture and implementation.