Back to Cookbook

Pipeline Paramedic

Stop guessing why your CI failed — get a diagnosis in seconds

Paste a failed CI log and get a root cause analysis, not a wall of red text. Catches the usual suspects — dependency resolution failures, flaky test patterns, Docker layer cache misses, permission errors, and timeout issues — then tells you exactly what to fix and where.

House RecipeWork1 min

INGREDIENTS

🐙GitHub

PROMPT

Create a skill called "Pipeline Paramedic". When I paste a CI/CD pipeline log, analyze it and identify the root cause of the failure. Classify it as one of: config error, flaky test, dependency issue, infrastructure problem, or timeout. For each finding, show the relevant log lines, explain what went wrong, and provide a specific fix (config change, code change, or workaround). If you detect flaky test patterns (race conditions, time-dependent assertions, shared state), flag them separately. Also suggest any pipeline optimizations you notice (missing caches, sequential steps that could parallelize, unnecessary steps).

How It Works

This skill reads CI/CD pipeline logs (GitHub Actions, GitLab CI, CircleCI,

Jenkins) and diagnoses failures the way a senior DevOps engineer would: by

pattern-matching against hundreds of known failure modes.

What You Get

  • Root cause identification from raw CI log output
  • Classification: config error, flaky test, dependency issue, infra problem, or timeout
  • Specific fix suggestions with code/config snippets
  • Detection of flaky test patterns (timing-dependent, order-dependent, shared state)
  • Pipeline optimization tips (caching, parallelization, unnecessary steps)

Setup Steps

  1. Copy your failed CI log output
  2. Paste it to your Claw and ask for a diagnosis
  3. Review the identified root cause and suggested fix
  4. Apply the fix or iterate with follow-up questions

Tips

  • Include the full log, not just the error line — context matters
  • Works with GitHub Actions, GitLab CI, CircleCI, Jenkins, and most CI systems
  • For flaky tests, paste logs from multiple runs to help identify the pattern
  • Ask your Claw to also suggest pipeline speed improvements while it's looking
Tags:#ci-cd#debugging#devops#automation