skills.rest
wangke19
wangke19COMMUNITY
Explore all skills by this author →
GitHub

prow-job-retry

Monitor Prow CI jobs, classify failures, and retry infrastructure failures.

1|1|Updated Apr 11, 2026
Search Tags:#monitoring#github#retry#ci#prow#backoff#failure-analysis
One-click install
npx skills add https://github.com/wangke19/my-claude-skills --skill prow-job-retry
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prow-job-retry
Source: https://github.com/wangke19/my-claude-skills/tree/main/plugins/ci/skills/prow-job-retry
Command: npx skills add https://github.com/wangke19/my-claude-skills --skill prow-job-retry

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, gh, gcloud, and includes scripts (resource) components.

What problem does it solve?

Automates monitoring of Prow CI jobs, analyzes failures to distinguish infrastructure vs code issues, and retries infrastructure failures with intelligent exponential backoff.

Core Features & Use Cases

  • Failure classification (INFRASTRUCTURE vs CODE) to minimize unnecessary retries.
  • Automatic retries for infrastructure failures with exponential backoff and jitter.
  • PR integration: posts retry comments and tracks new job IDs for end-to-end retry workflows across presubmit, periodic, and rehearsal jobs.

Quick Start

Start monitoring a Prow job by invoking the monitor script with the bucket path, PR repository, and PR number.

Frequently Asked Questions about prow-job-retry

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

FAQPage Schema
How do I automatically retry flaky Prow CI jobs?▼

You can automatically retry flaky Prow CI jobs by running a monitor script that reads prowjob.json and status files from GCS, classifies failures, and posts PR comments to trigger retries with exponential backoff.

How does failure classification work for Prow CI infrastructure issues?▼

Failure classification for Prow CI issues distinguishes between INFRASTRUCTURE and CODE failures by analyzing job status files, ensuring only infrastructure failures are retried to minimize unnecessary automated actions.

Does Prow CI retry automation work with periodic and rehearsal jobs?▼

Yes, Prow CI retry automation supports end-to-end retry workflows across presubmit, periodic, and rehearsal jobs by tracking new job IDs after posting PR comments to trigger the retry process.

Do I need jq and gcloud to monitor Prow jobs from a GCS bucket?▼

Yes, you need jq, gcloud, and gh installed to monitor Prow jobs, because the automation requires these tools to read prowjob.json from GCS and interact with GitHub PRs.

Why does my Prow job retry use exponential backoff with jitter?▼

Prow job retries use exponential backoff with jitter to intelligently space out automated retry attempts for infrastructure failures, preventing sudden bursts of load on the CI system.

Related Skills

Explore More Software Engineering Skills→
openshift-engopenshift-eng

Prow Job Artifact Search

Search and retrieve Prow CI job artifacts from the test-platform-results GCS bucket.

Official
Intermediate
rahulrajaramrahulrajaram

analyze-conversation

Analyze completed AI conversations to identify anti-patterns and generate markdown reports.

Community
Advanced
testdoubletestdouble
198

post-code-review-to-pr

Post code review results to GitHub PRs as comments or formal reviews.

Official
Advanced
openshift-engopenshift-eng
7

microshift-ci:prow-job

Analyzes Prow build logs to identify failing CI steps and classify root causes for MicroShift workflows.

Official
Advanced