Execute documented instructions against real environments and report verification results.
npx skills add https://github.com/stacklok/docs-website --skill test-docs
Please help me install this Agent Skill. Skill: test-docs Source: https://github.com/stacklok/docs-website/tree/main/.claude/skills/test-docs Command: npx skills add https://github.com/stacklok/docs-website --skill test-docs
This Skill helps prevent inaccurate or outdated documentation by running ToolHive documentation instructions in a real environment and identifying broken examples before users encounter them.
Use the test-docs skill to verify the ToolHive vault integration tutorial and report whether every documented step passes.
High-intent search queries and answers about installing and using this skill.
You can validate documentation by executing documented instructions against a live Kubernetes environment, tracking sequential execution, and classifying failures with evidence-based reports to confirm whether tutorials work as expected.
Documentation testing prevents broken CLI tutorials by running documented commands against real environments, extracting expected results, and reporting verification failures to maintainers before users encounter outdated or inaccurate instructions.
Yes, you need kubectl, helm, kind, thv, jq, and curl installed to perform deterministic prerequisite checks and execute documented instructions against a live environment for ToolHive tutorials.
You can identify failure sources by executing documented steps against a real environment and applying evidence-based failure classification to distinguish documentation errors from ToolHive bugs or environment issues.
Yes, you can verify CLI guides and Kubernetes manifests by extracting commands and manifests from documentation, executing them sequentially, and generating test reports with pass or fail evidence and recommendations.
The best way to test documentation accuracy is running documented instructions against real environments with deterministic prerequisite checks, sequential execution tracking, and evidence-based failure classification to identify broken examples.
missBergRun end-to-end tests for Envoy Gateway and AI Gateway on Kubernetes clusters.
babywormClassify LLM-generated code review findings into severity-labeled reports with reproducible evidence.
salmanparachaDeploy PostgreSQL clusters on Kubernetes with CloudNativePG and automated failover.
nguyenphpRun linters, type checkers, and security audits across Node.js and Python projects.
