ci-monitor

Monitor GitHub Actions failures and classify them for automated fixes or issue creation.

1|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/mattbutlerengineering/mattbutlerengineering --skill ci-monitor-mattbutlerengineering
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ci-monitor
Source: https://github.com/mattbutlerengineering/mattbutlerengineering/tree/main/.claude/skills/ci-monitor
Command: npx skills add https://github.com/mattbutlerengineering/mattbutlerengineering --skill ci-monitor-mattbutlerengineering

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you detect failing GitHub Actions on the main branch and quickly restore developer velocity by fixing straightforward errors or escalating complex ones.

Core Features & Use Cases

  • Main CI monitoring: Lists recent main-branch runs, identifies failing workflows, and inspects failed logs.
  • Failure triage and handling: Classifies failures into simple (lint/type/test/prisma-gen/small test fix) versus complex (flaky tests, infra issues, multi-file refactors, dependency conflicts).
  • Auto-fix and escalation: Auto-fixes simple failures by running the agent with a tight budget and creates labeled GitHub issues for complex failures; also monitors open PRs created by the agent and attempts fixes for failing checks.

Quick Start

Ask the agent to run the CI monitor by invoking /ci-monitor and let it check main branch CI, fix simple breakages, and open issues for complex failures.

Frequently Asked Questions about ci-monitor

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

FAQPage Schema
How do I automatically fix failing GitHub Actions workflows on my main branch?

Automated CI monitoring inspects failed GitHub Actions logs, classifies failures by severity, and runs targeted agent remediation for simple errors like lint or Prisma generation issues, escalating complex failures to labeled GitHub issues.

What types of CI failures can be auto-fixed versus escalated?

Simple failures including lint, type, test, and Prisma generation errors are auto-fixed with a tight agent budget. Complex failures like flaky tests, infra issues, dependency conflicts, and multi-file refactors are escalated to labeled GitHub issues.

Can I monitor and fix failing checks on agent-created pull requests?

Yes, CI monitoring tracks open pull requests created by the agent and attempts automated fixes for their failing checks using the same log inspection and targeted recovery workflow applied to main branch runs.

Does CI monitoring require the gh CLI to inspect GitHub Actions logs?

Yes, CI monitoring requires the gh CLI to list recent workflow runs, view failed logs, and check for existing ci-fix issues before initiating automated remediation or creating new issues for complex failures.

What are the limitations of automated GitHub Actions failure remediation?

Automated remediation is limited to simple failures like lint, type, and Prisma generation errors under strict budget caps. It will not blindly retry runs, and complex failures involving flaky tests, infra issues, or dependency conflicts are escalated to issues.