بنقرة واحدة
trigger-pr-jobs
Identify and retrigger missing or failed required Prow status checks for a GitHub pull request.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Identify and retrigger missing or failed required Prow status checks for a GitHub pull request.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Analyze the root cause inside a failed prowjob. Use when user mentions to analyze a prowjob failure or adds a prowjob url
Analyze root causes for all the recent ci-failures. Use when user mentions to analyze recent ci failures
Analyze all failed builds for a GitHub pull request. Use when user provides a GitHub PR URL and wants to understand why builds are failing.
Analyze the root cause inside a failed prowjob when a user mentions a failure or adds a prowjob URL.
Analyze root causes for all the recent CI failures (failing prowjobs that have not reached the testing stage).
Analyze all failed builds for a GitHub pull request and identify root causes.
| name | trigger-pr-jobs |
| description | Identify and retrigger missing or failed required Prow status checks for a GitHub pull request. |
| argument-hint | <pr-url> |
| allowed-tools | Bash(gh api *), Bash(gh pr checks *), Bash(gh pr comment *), Bash(gh pr view *) |
This skill identifies required status checks that are either missing (no result) or failed on a GitHub pull request, then posts /test commands to retrigger them.
It works for any repo in the kubevirt GitHub org that uses Prow.
Parse the PR URL from $ARGUMENTS to get owner, repo, and pr_number. Validate that all three values were successfully extracted and are non-empty. If the URL is malformed or any component is missing, report the error to the user and stop.
gh api repos/{owner}/{repo}/pulls/{pr_number} --jq '.base.ref'
gh api repos/{owner}/{repo}/branches/{branch}/protection/required_status_checks --jq '.contexts[]'
If this returns a 404 (branch protection not configured), treat it as having no required checks. Tell the user that no required status checks are configured for the target branch and stop.
Get the status of all checks reported on the PR:
gh pr checks {pr_number} --repo {owner}/{repo} --json name,state,bucket
This returns a JSON array of objects containing the check name and its current state (e.g., SUCCESS, FAILURE, PENDING).
Compare the required checks against the checks present on the commit. A check needs triggering if it is:
fail statusOnly trigger Prow jobs. Filter to job names starting with pull-. Non-Prow checks cannot be triggered via /test and must be excluded:
dco, tide — Prow tide/plugin statuses, not jobscoverage/coveralls — third-party servicecheck-vep) — triggered by GitHub, not ProwSourcery review)If there are no checks to trigger, tell the user and stop.
Post a single comment on the PR with one /test {job_name} line per check that needs triggering:
gh pr comment {pr_number} --repo {owner}/{repo} --body "/test job-1
/test job-2
/test job-3"
Always combine all /test lines into one comment to avoid noise on the PR.
Before posting /test commands, consider whether retriggering is likely to help:
analyze-build or test-failure-rate first instead of wasting CI resourcesWhen there are many failures, suggest running test-failure-rate on one of the failed build URLs first to quickly assess whether the failures are flaky. If all failures are known flakes, retrigger confidently. If most are likely PR-related, warn the user before retriggering.
Present a summary table to the user showing:
/test commands were posted, with a link to the comment