pages-build-shepherd

Diagnose and repair failed GitHub Pages build deployments via run logs and GitHub CLI.

3|2|Updated May 12, 2026
One-click install
npx skills add https://github.com/kriscendobot/garden --skill pages-build-shepherd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pages-build-shepherd
Source: https://github.com/kriscendobot/garden/tree/main/skills/pages-build-shepherd
Command: npx skills add https://github.com/kriscendobot/garden --skill pages-build-shepherd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill resolves failed GitHub Pages deployments by automatically classifying the failure type and executing the necessary recovery procedure, ensuring your documentation site remains live without manual intervention.

Core Features & Use Cases

  • Automated Failure Classification: Distinguishes between transient deployment flakes, content-related build errors, and configuration impasses.
  • Intelligent Recovery: Automatically triggers re-runs for transient issues or applies surgical fixes to the docs source for build errors.
  • Use Case: When a push to the main2 branch causes a Pages build failure, the shepherd identifies the root cause, performs a clean worktree-based fix if necessary, and verifies the deployment returns to a green state.

Quick Start

Use the pages-build-shepherd skill to investigate and repair the latest failed pages-build-deployment run for the current repository.

Frequently Asked Questions about pages-build-shepherd

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

FAQPage Schema
How do I automate GitHub Pages deployment troubleshooting and recovery?

Automate GitHub Pages deployment troubleshooting by analyzing run logs to classify failures, applying worktree-based code fixes for build errors, and re-running transient flakes via the GitHub CLI to verify a green state.

Why does my GitHub Pages build fail after pushing to the main branch?

GitHub Pages build failures after a push are classified as transient deployment flakes, content-related build errors, or configuration impasses, diagnosed by analyzing the run logs to determine the necessary recovery procedure.

What's the best way to fix a failed GitHub Actions workflow for Pages without manual intervention?

Fix failed GitHub Actions workflows for Pages automatically by classifying the failure type, applying surgical fixes to the docs source in an isolated worktree, and verifying deployment success through the GitHub CLI.

Do I need write access to manage GitHub Actions runs and repair Pages deployments?

Yes, repairing Pages deployments requires read and write access to the repository's Pages source branch and the ability to manage GitHub Actions runs to execute recovery procedures and verify deployment success.

Can a CI-CD tool distinguish between transient GitHub Pages flakes and actual build errors?

Yes, the tool distinguishes between transient deployment flakes, content-related build errors, and configuration impasses by analyzing run logs, triggering re-runs for transient issues or applying surgical fixes for build errors.

When should I not use automated recovery for GitHub Pages deployments?

Automated recovery is limited when failures stem from complex configuration impasses it cannot resolve surgically, requiring manual intervention for issues beyond transient flakes or content-related build errors.