CI Doctor
Debug failing GitHub Actions without scrolling 10,000 log lines. Reads the logs from the first failing step, pinpoints the root cause (missing secret, dependency mismatch, YAML syntax, permissions), and hands you a copy-paste fix, then offers to open the PR. GitHub CLI required.
- Identify the root cause of failing GitHub Actions or GitLab CI jobs
- Automatically fix YAML syntax errors in workflow configuration files
- Debug dependency and lockfile mismatches that break build environments
$12
· or 60 creditsSecure checkout via Stripe
Included in download
- Identify the root cause of failing GitHub Actions or GitLab CI jobs
- Automatically fix YAML syntax errors in workflow configuration files
- terminal, file_read, file_write automation included
- Includes example output and usage patterns
Sample input
Use $ci-doctor to figure out why my build is failing and help me fix the pipeline.
Sample output
Root Cause: The 'build' job failed because the 'production' environment requires a 'NPM_TOKEN' secret which is not configured. Evidence: 401 Unauthorized - GET https://registry.npmjs.org/ Fix: Add NPM_TOKEN to your repository secrets. PR Option: I can update the YAML to use the secret now.
Debug failing GitHub Actions without scrolling 10,000 log lines. Reads the logs from the first failing step, pinpoints the root cause (missing secret, dependency mismatch, YAML syntax, permissions), and hands you a copy-paste fix, then offers to open the PR. GitHub CLI required.
$12
· or 60 creditsSecure checkout via Stripe
Also available in a bundle
Included in download
- Identify the root cause of failing GitHub Actions or GitLab CI jobs
- Automatically fix YAML syntax errors in workflow configuration files
- terminal, file_read, file_write automation included
- Includes example output and usage patterns
- Instant install
Sample input
Use $ci-doctor to figure out why my build is failing and help me fix the pipeline.
Sample output
Root Cause: The 'build' job failed because the 'production' environment requires a 'NPM_TOKEN' secret which is not configured. Evidence: 401 Unauthorized - GET https://registry.npmjs.org/ Fix: Add NPM_TOKEN to your repository secrets. PR Option: I can update the YAML to use the secret now.
About This Skill
Automated CI/CD Troubleshooting
CI Doctor gets your pipeline back to green without the manual slog of reading thousands of build-log lines. It parses the logs from the first failing step, identifies the root cause, and proposes a targeted fix. Built for GitHub Actions, with detection for GitLab CI and CircleCI when their config files or log markers appear.
How it works
Unlike a generic AI prompt, CI Doctor follows a structured diagnostic framework. For GitHub Actions, it uses the GitHub CLI (`gh`) to pull logs from the failing step rather than the final summary, then classifies the failure into a specific category like dependency error, permission issue, missing secret, or config syntax, so the fix is surgical instead of guessed. For GitLab CI and CircleCI, it recognizes the provider from your config and log markers and diagnoses from the evidence you provide.
Features
- Deep Log Analysis: For GitHub Actions, automatically fetches and parses logs using `gh run view --log-failed`.
- Provider Detection: Fully automated on GitHub Actions; detects GitLab CI (`.gitlab-ci.yml`) and CircleCI (`.circleci/config.yml`) and diagnoses from the config and logs you supply.
- Root Cause Identification: Distinguishes cascading errors from the actual source of the failure.
- Safe Implementation: Proposes fixes and creates pull requests only after developer confirmation.
- Context Awareness: Inspects package manifests and lockfiles so fixes are compatible with your tech stack.
The Output
You receive a concise report containing a one-sentence root cause, the specific evidence from the logs, a copy-pasteable fix (code or YAML), and an offer to automate the pull request creation.
Use Cases
- Identify the root cause of failing GitHub Actions or GitLab CI jobs
- Automatically fix YAML syntax errors in workflow configuration files
- Debug dependency and lockfile mismatches that break build environments
- Identify and resolve missing secrets or environment variables in pipelines
- Generate and push a PR with a confirmed fix to unblock the main branch
Known Limitations
Requires GitHub CLI (gh) installed and authenticated (gh auth login). Full automation on GitHub Actions; GitLab CI and CircleCI are detected and diagnosed from supplied config and logs. Fix PRs require user confirmation before creation. Cannot fix self-hosted runner configuration issues. Does not modify workflow settings in GitHub UI (e.g., branch protection rules).
How to Install
mkdir -p ~/.claude/skills && curl -sL https://www.agensi.io/api/install/ci-doctor -o /tmp/ci-doctor.zip && unzip -o /tmp/ci-doctor.zip -d ~/.claude/skills && rm /tmp/ci-doctor.zipFree skills install directly. Paid skills require purchase - use the download button above after buying.
Reviews
No reviews yet - be the first to share your experience.
Only users who have downloaded or purchased this skill can leave a review.
Early access skill
Be the first to review this skill.
Only users who have downloaded or purchased this skill can leave a review.
Security Scanned
Passed automated security review
Permissions
Allowed Hosts
File Scopes
CI Doctor needs terminal access to run the GitHub CLI (`gh`) and git commands. These commands fetch workflow run status, download failure logs, and (with user confirmation) create fix PRs. The skill never runs destructive git operations (force push, reset) and never creates PRs without explicit user approval. File read access allows CI Doctor to examine workflow YAML files (`.github/workflows/*.yml`), dependency manifests (`package.json`, `requirements.txt`, `go.mod`), and error logs to understand the CI environment and identify root causes like missing dependencies, syntax errors, or misconfigured secrets. File write access is used only when the user explicitly confirms "create a fix PR." The skill then creates a new branch, modifies the workflow file, commits the change, and opens a PR. All changes are previewed to the user before any write occurs. The skill never writes to files without user consent. Network access fetches workflow run data from `api.github.com` as a fallback if the GitHub CLI is not installed. These are read-only API calls authenticated via the user's GitHub token (provided by the user or read from local git config). No data is sent to external servers beyond GitHub's API. No browser or environment variable access is required.
Works with Claude Code and Codex. Requires GitHub CLI (gh) installed and authenticated. Supports GitHub Actions only initially; GitLab CI and CircleCI planned.
Creator
JustHandled Labs builds focused agent skills for the work nobody wants to do by hand. Each one is a single repeatable job done well: catching the security and data mistakes that quietly ship, keeping docs and tests honest, gating the commands an agent is about to run, sharpening writing, and handling the founder chores around launches, outreach, and brand setup. Not generic AI productivity. Specific workflows that are easy to run, review, and repeat. Maintained by H.J. Westerfield, with a background in communications, editing, project coordination, customer support, and practical AI systems. Tools for people who want useful automation without theatrical complexity.
Frequently Asked Questions
Learn More About AI Agent Skills
More Premium Skills
Multi-Agent Orchestration Master Library
Transform Claude Code into a coordinated multi-agent system. Battle-tested tmux orchestration patterns, YAML task queues, event-driven communication, and parallel worker management for 8+ agents.
diagnosing-rag-failure-modes
RAG fails quietly. It retrieves documents, returns confident-looking answers, and misses the question entirely — because the question required connecting facts across documents, reasoning about sequence, or tracing causation. This skill gives you a five-question diagnostic checklist that classifies any failing query as either RAG-safe or structurally RAG-incompatible, then maps it to the specific failure pattern and the architectural fix that resolves it.
sast-configuration
Automate the setup and optimization of Semgrep, SonarQube, and CodeQL for high-signal security testing.
designing-hybrid-context-layers
Architects the right retrieval strategy for every query — teaching your agent when to use RAG, a knowledge graph, or a temporal index instead of defaulting to vector search for everything.