Using with AI
AI coding assistants can help you get the most out of Markuplint — from understanding warnings to writing configuration files.
Understanding warnings
When Markuplint reports a violation, you can paste the error message into an AI assistant to get an explanation and suggested fix. Each violation includes a rule name (e.g., attr-duplication) that the AI can look up.
The attribute name is duplicated Markuplint(attr-duplication) [Ln 11, Col 45]
Ask your AI assistant:
- "What does this Markuplint warning mean?"
- "How do I fix this HTML to resolve
permitted-contents?"
Writing configuration files
AI assistants can generate Markuplint configuration files tailored to your project. Describe your setup and let the AI create the config:
- "Create a
.markuplintrcfor a React project" - "Add a rule to enforce BEM class naming in my Markuplint config"
- "Configure Markuplint to allow the
data-testidattribute"
AI coding agents
AI coding agents such as Claude Code, Cursor, and GitHub Copilot can run Markuplint directly in your development workflow:
- Lint generated code — Ask the agent to run
npx markuplintafter generating HTML - Auto-fix violations — The agent can interpret Markuplint's output and apply fixes
- Set up projects — The agent can create a
.markuplintrcfile directly for your framework
npx markuplint --init is interactive and requires manual input. AI agents should write the configuration file directly instead. See Beyond HTML for the parser and spec packages to include.
For AI agent developers: a dedicated SKILL file for Markuplint is planned to help agents use Markuplint more effectively.
Next steps
- Getting Started — Install the VS Code extension and start linting
- Beyond HTML — Set up parsers for your framework
- Configuration — Learn about all available configuration options