yaml-schema-lint GitHub Action

A composite GitHub Action that lints YAML files against JSON schemas using yaml-schema-lint and creates a GitHub Check with inline annotations and a markdown summary

Usage

yaml-schema-lint:
  runs-on: ubuntu-latest
  permissions:
    contents: read
    checks: write
    pull-requests: read
  steps:
    - uses: actions/checkout@v4

    - uses: X-Guardian/yaml-schema-lint-action@v1.0.0
      with:
        patterns: "'**/*.yml' '**/*.yaml'"

Inputs

Input	Description	Required	Default
patterns	YAML file paths or glob patterns (space-separated)	Yes	—
ignore	Glob patterns to exclude (space-separated)	No	**/node_modules/**
settings-path	Path to a VS Code settings.json with yaml.schemas	No	.vscode/settings.json
no-schema-store	Disable fetching schemas from schemastore.org	No	false
version	yaml-schema-lint version to install (ignored when binary is set)	No	1.1.0
binary	Path to a local yaml-schema-lint binary (skips npx install)	No	—
node-version	Node.js version to use	No	24
check-name	Name shown in the GitHub Checks tab	No	YAML Schema Lint Report
fail-on-error	Fail the check when errors are found	No	true
fail-on-warning	Fail the check when warnings are found	No	true
only-pr-files	Only annotate files changed in the pull request	No	true
markdown-report-on-step-summary	Write a markdown summary to the step summary	No	true

Permissions

The job using this action needs the following permissions:

permissions:
  contents: read # to checkout the repository
  checks: write # to create the Check Run
  pull-requests: read # to list PR-changed files (when only-pr-files is true)

How it works

1. Setup Node.js using actions/setup-node
2. Install and run yaml-schema-lint, producing a JSON report
3. Create a GitHub Check Run via the Checks API with:
   • A conclusion of success or failure based on the fail-on-error / fail-on-warning inputs
   • Inline annotations on the PR diff for each diagnostic
   • A markdown summary table written to the GitHub Actions step summary
   • Annotations are batched (max 50 per API call) to handle large reports