runegard

Parse Kubernetes runbooks into executable decision trees for automated execution.

5|Updated Mar 8, 2026

Search Tags:#automation #llm #claude #kubernetes #approval #runbooks #continual-learning

Ready to run

npx skills add https://github.com/mvilrokx/runegard --skill runegard

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: runegard
Source: https://github.com/mvilrokx/runegard/tree/main
Command: npx skills add https://github.com/mvilrokx/runegard --skill runegard

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Automates the parsing of Kubernetes runbooks and their step-by-step execution on live clusters.

Core Features & Use Cases

Automated runbook parsing: converts Markdown-based procedures into a deterministic execution plan.
Safe execution: executes diagnostic and remediation steps with an approval gate before mutations.
Continual learning: analyzes failures and patches the skill using an RLM loop to improve future runs.

Quick Start

Run a runbook by providing its path to let RuneGard parse, build a decision tree, and execute it against your Kubernetes cluster (dry-run option available).

Frequently Asked Questions about runegard

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate Kubernetes runbook execution from Markdown files?▼

Automating Kubernetes runbook execution involves parsing Markdown procedures into executable decision trees. RuneGard converts these runbooks into deterministic plans for automated diagnosis and remediation against live clusters. It supports Markdown and other formats with a dry-run option.

How do Kubernetes runbooks learn from past execution failures?▼

Kubernetes runbooks learn from past failures through a continual learning loop that analyzes execution traces. RuneGard uses an RLM improvement loop to patch the skill after analyzing failures, ensuring automated runbook execution improves for future runs.

Can I execute remediation steps on a Kubernetes cluster with human approval gates?▼

Executing remediation steps with human approval gates is supported by RuneGard. It applies an approval gate before any mutation steps, enabling safe execution of diagnostic and remediation actions on your live Kubernetes cluster.

Do I need a Claude-compatible parser to automate Kubernetes runbooks?▼

A Claude-compatible parser is required to automate Kubernetes runbooks with RuneGard. The execution engine uses this parser alongside a defined step taxonomy to build deterministic decision trees and execute runbooks against live clusters.

What is the best way to parse Markdown runbooks into executable Kubernetes operations?▼

The best way to parse Markdown runbooks into executable operations is using a parser that converts procedures into deterministic execution plans. RuneGard builds a decision tree from Markdown runbooks to automate diagnosis, remediation, verification, and escalation.

Does automated runbook execution support dry-run verification before applying changes?▼

Automated runbook execution supports dry-run verification before applying changes. RuneGard allows you to run a runbook by providing its path to parse and build a decision tree, testing it against your Kubernetes cluster with a dry-run option before mutations.