One-click install
npx skills add https://github.com/openshift-eng/edge-tooling --skill lvms-ci-doctor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lvms-ci:doctor
Source: https://github.com/openshift-eng/edge-tooling/tree/main/plugins/lvms-ci/skills/doctor
Command: npx skills add https://github.com/openshift-eng/edge-tooling --skill lvms-ci-doctor

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gsutil, python3, bash, and includes scripts (resource) components.

What problem does it solve?

LVMS release teams need fast, readable insight into why periodic CI jobs are failing, but manually correlating job failures across many releases and artifacts is time-consuming and error-prone.

Core Features & Use Cases

  • Release-scoped failure analysis: Takes a comma-separated list of release versions (including main), then analyzes only the periodic job failures for each release.
  • Deterministic artifact collection + HTML reporting: Uses deterministic scripts to fetch/download job artifacts, aggregates results, and generates a single self-contained HTML report file.
  • Per-job root cause generation: Uses LLM agents only for per-job analysis after deterministic preparation, producing consistent per-job writeups for final aggregation.

Quick Start

Run /lvms-ci:doctor 4.20,4.21,main to analyze LVMS periodic CI failures for those releases and produce an HTML summary report.

Frequently Asked Questions about lvms-ci:doctor

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

FAQPage Schema
How do I analyze LVMS CI failures across multiple releases?

To analyze LVMS CI failures across releases, aggregate periodic Prow job results into a single HTML report. Provide a comma-separated list of release versions to generate a timestamped, self-contained summary of per-job root causes.

What is the best way to generate an HTML report for Prow periodic job failures?

Generating an HTML report for Prow periodic job failures requires deterministic artifact collection followed by LLM-based per-job analysis. This approach aggregates downloaded artifacts into a single actionable summary file for release engineering workflows.

Do I need gsutil and python3 to collect CI artifacts for root cause analysis?

Yes, you need gsutil, python3, and bash to perform deterministic CI artifact collection and aggregation. These dependencies fetch Prow job artifacts and execute the scripts required to prepare data for root cause analysis.

Can I run release engineering failure analysis for the main branch alongside specific versions?

Yes, you can include the main branch alongside specific versions for failure analysis. Specify a comma-separated list like '4.20,4.21,main' to aggregate periodic CI job failures across all requested releases.

How does per-job root cause analysis work for periodic CI failures?

Per-job root cause analysis works by using deterministic scripts to fetch and prepare artifacts, then applying LLM agents only for the per-job analysis phase. This produces consistent, actionable writeups that are aggregated into a final HTML report.

Where does the aggregated CI failure summary report get saved?

The aggregated CI failure summary report saves as 'lvm-operator-ci-doctor-report.html' under a computed timestamped work directory. This deterministic output location ensures engineers can reliably locate the generated HTML summary file.