pr-health
Check PR health status (conflicts, CI, approval) and get actionable next steps
Best use case
pr-health is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Check PR health status (conflicts, CI, approval) and get actionable next steps
Teams using pr-health should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/pr-health/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How pr-health Compares
| Feature / Agent | pr-health | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Check PR health status (conflicts, CI, approval) and get actionable next steps
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# PR Health Skill Check the health of a pull request including merge conflicts, CI status, and approval status. ## Commands ### Check PR Health ```bash # Check PR for current branch toolkit pr health # Check specific PR toolkit pr health 123 # Output as JSON toolkit pr health --json # Specify repository toolkit pr health --repo owner/repo ``` ### Get Workflow Logs ```bash # Get all logs for a run toolkit pr logs <run-id> # Get only failed job logs toolkit pr logs <run-id> --failed-only # Get logs for specific job toolkit pr logs <run-id> --job "build" ``` ### Detect PR ```bash # Find PR for current branch toolkit pr detect # Output as JSON toolkit pr detect --json ``` ## Output Format The health command outputs a structured report: ``` ## PR Health Report: #123 **URL:** https://github.com/owner/repo/pull/123 ### Status: UNHEALTHY (2 issues) ### Merge Conflicts: UNHEALTHY - Branch has merge conflicts with base - Conflicting file: src/lib/parser.ts To investigate: \`\`\`bash git fetch origin main && git merge origin/main \`\`\` ### CI Status: FAILED - Job "test" - FAILED - Run ID: 12345678 To investigate: \`\`\`bash toolkit pr logs 12345678 --failed-only \`\`\` ### Approval: APPROVED - claude-code-review[bot]: APPROVED ### Next Steps 1. Resolve merge conflicts 2. Fix CI failures ``` ## Status Values - **HEALTHY**: No issues - **UNHEALTHY**: Issues need to be addressed - **PENDING**: Waiting for something (CI running, reviews needed) ## Troubleshooting ### "No PR found for current branch" Create a PR first or specify a PR number: ```bash gh pr create # or toolkit pr health 123 ``` ### CI logs are empty The workflow run may not have generated logs yet. Wait for the run to complete. ### Merge conflict detection is inaccurate Ensure you have fetched the latest from origin: ```bash git fetch origin main ```
Related Skills
toolkit-recall
Search past decisions, prior research, conversation history, monorepo docs, and fetched web pages. Use toolkit recall search to find context from previous work, and toolkit fetch to save web pages for future search.
zod-patterns
Zod schema validation, type-safe development, and strict TypeScript patterns. When user works with Zod, validates data, creates schemas, handles form validation, mentions z.object/z.string patterns, needs runtime validation, type-safe code, or strict TypeScript configuration.
zellij-helper
Zellij terminal multiplexer for session management, layouts, and pane operations When user mentions Zellij, terminal multiplexer, zellij commands, sessions, panes, layouts, or tabs
worktree-workflow
Git worktree workflow for isolated feature development and PR creation When user starts new work, needs to switch contexts, or wants parallel development
vite-react-helper
Vite + React for fast modern web development - build config, HMR, hooks, state management, and performance patterns When user works with Vite, React, creates components, manages state, uses hooks, or configures Vite builds
version-management
Use when asking about version management, Renovate annotations, versions.ts patterns, or pinning image/chart versions.
typst-authoring
This skill should be used when the user asks to "write Typst", "create a Typst document", "format in Typst", "convert to Typst", "Typst syntax", "Typst template", "Typst math", "Typst table", or works with .typ files. Provides comprehensive Typst markup, scripting, math, layout, and styling reference for authoring documents. Also use proactively when generating .typ output files (e.g., in deep-research reports).
typescript-helper
TypeScript development guidance for type systems and tooling When user works with .ts or .tsx files, mentions TypeScript, or encounters type errors
torvalds-deployment
Use when asking about adding services, createXxxDeployment patterns, or homelab deployment conventions. Services use per-namespace charts.
terraform-helper
Terraform and OpenTofu infrastructure as code - HCL, providers, modules, state management, and CLI operations When user works with .tf files, mentions Terraform, OpenTofu, tofu, HCL, infrastructure as code, or tf commands
terminal-concepts
Comprehensive guide for building CLI and TUI applications - terminal internals, design principles, and battle-tested patterns When building CLI/TUI apps, implementing argument parsing, handling terminal input/output, escape codes, buffering, signals, or asking about terminal development concepts
talos-helper
Talos Linux cluster administration using talosctl When user mentions Talos, talosctl, or Talos cluster operations