ci-watchdog

Monitor GitHub PR CI checks and automatically fix failures until all pass.

4.6k|378|Updated Jun 28, 2024
One-click install
npx skills add https://github.com/latitude-dev/latitude-llm --skill ci-watchdog
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ci-watchdog
Source: https://github.com/latitude-dev/latitude-llm/tree/main/.agents/skills/ci-watchdog
Command: npx skills add https://github.com/latitude-dev/latitude-llm --skill ci-watchdog

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Continuously monitor PR CI checks and automatically fix failures until all checks pass.

Core Features & Use Cases

  • Continuous monitoring of PR CI checks across GitHub Actions and other CI providers.
  • Automatic failure detection, log-based diagnosis, and fixes with commit and push.
  • Real-world use: keep a PR green after patches by looping fixes until CI passes.

Quick Start

Watch the PR's CI checks and automatically apply fixes until all checks pass.

Frequently Asked Questions about ci-watchdog

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

FAQPage Schema
How do I automatically fix failing CI checks on a GitHub pull request?

You can automatically fix failing CI checks by continuously monitoring the GitHub PR workflows, diagnosing the specific failure from logs, and looping code fixes with commits until green. This Skill automates that entire diagnosis and patching loop.

Can I monitor CI workflows across providers other than GitHub Actions?

Yes, continuous monitoring of PR CI checks works across GitHub Actions and other CI providers. The Skill detects failures, diagnoses logs, and pushes fixes regardless of the specific CI system running the checks.

What is the best way to keep a pull request green after submitting patches?

The best way to keep a PR green after patches is to loop automated fixes until CI passes. The Skill continuously watches CI checks, diagnoses log failures, and pushes code changes repeatedly to maintain green CI across runs.

How does automated CI failure diagnosis work from logs?

Automated CI failure diagnosis works by watching check runs, extracting errors from CI logs, generating targeted code changes, and pushing commits. It repeats this loop, verifying each fix until all CI checks pass successfully.

What are the limitations of automatically fixing CI failures from logs?

Limitations of automatically fixing CI failures from logs include dependency on clear error messages for diagnosis and the risk of looping indefinitely on unresolvable issues. It works best when log outputs explicitly indicate the failing code or configuration.

Do I need GitHub Actions configured to use automated CI monitoring?

No, GitHub Actions is not strictly required. The Skill applies to PR workflows across GitHub Actions and other CI systems, watching for failures, diagnosing issues from logs, and looping fixes until all checks pass.