n8n-nodes-markupai

This is an n8n community node that integrates Markup AI — a content guardian platform for ensuring brand-compliant, consistent, and engaging content.

Markup AI provides AI-powered content analysis through specialized agents (for example terminology, claims, focus, and AI voice detection). Integrate agent-based quality checks and automated improvements directly into your n8n workflows.

Installation · Operations · Credentials · Usage · Resources · Development

Installation

Community Nodes (Recommended)

1. In n8n, go to Settings > Community Nodes
2. Select Install
3. Enter @markupai/n8n-nodes-markupai in the package field
4. Click Install

For detailed instructions, visit the n8n community nodes installation guide.

Manual Installation

To install manually, run the following command in your n8n installation directory:

npm install @markupai/n8n-nodes-markupai

Operations

The Markup AI node currently supports one operation:

Run Agent

Run the Markup AI Style Agent on input text.

• Execution model: Calls POST /agents/{id}/run and polls GET /agents/workflows/{id} until the workflow reaches a terminal state (completed, failed, timed_out, or cancelled) or the configured timeout elapses.
• Gating: The node fails fast if your organization has style_agent: "disabled" in /style-agent/config.

Returns (success):

• workflow_id, status, started_at, completed_at, duration_seconds
• document_ref (echoed back from your input, when provided)
• result (agent output payload)
• issue_counts (total, high, medium, low computed from result.issues)
• html_report (rendered HTML report of issues and workflow metadata)

Returns (failure with "Continue On Fail"): When the node is configured with Continue On Fail and the run errors, the item output is simply { error: "<message>" }. Without Continue On Fail, errors are thrown as n8n node-operation errors and the workflow stops.

Credentials

To use this node, you'll need a Markup AI API account.

Setting up credentials:

1. Sign up for a Markup AI account
2. Get your API key from your account settings
3. In n8n, add new credentials:
   • Go to Credentials > New
   • Search for Markup AI API
   • Enter your API key
   • Click Save

Usage

Configuration Options

The Run Agent operation supports the following configuration:

| Option | Type | Required | Default | Description | | ------------------------------------------- | ------------------ | -------- | -------- | ----------------------------------------------------------------------------------------------------------- | | Agent | Single-select | Yes | - | Select the agent to run. Currently only style_agent is exposed. | | Style Guide | Single-select | No | - | Style agent style guide. Loaded from GET /style-agent/style-guides. Defaults to your default style guide. | | Content | String (multiline) | Yes | - | Document text to analyze. | | Additional Options → Document Name | String | No | "" | Human-readable name of the document being analyzed. | | Additional Options → Document Reference | String | No | "" | Caller-supplied identifier (e.g. CMS page ID) echoed back in the result for tracking. | | Additional Options → Timeout (Ms) | Number | No | 120000 | Maximum time to wait while polling workflow status before returning a timeout error. |

Organization and content profile are auto-detected from the API key — there is no input for them.

Example Workflows

Content Quality Gate

1. Trigger: New document added to Google Drive
2. Markup AI → Run Agent: Select one or more compliance-focused agents
3. Condition: Check status and evaluate returned issues from result
4. If passed: Approve and publish
5. If failed: Send notification to content team

Automated Content Improvement

1. Trigger: Webhook receives content for publication
2. Markup AI → Run Agent: Analyze with selected agents (for example terminology + generic claims)
3. Send improved version to CMS for publication

Common Use Cases

• Content Compliance & Consistency: Ensure all company content — from blogs and web pages to documentation and marketing materials — aligns with brand and style guidelines while maintaining a consistent tone.

• Automated Quality Gates: Ensure content meets standards automatically before publication. Monitor content changes in real-time and prevent non-compliant content from going live.

• Quality at the Source: Create and maintain high-quality content right where it's produced. Proactively monitor consistency and compliance across CMS platforms like Google Drive and beyond.

• Early Issue Detection: Automatically identify and flag potential issues or non-compliant content within workflows before they become problems.

• Streamlined Content Compliance: Minimize manual reviews by automating content compliance checks and review processes throughout your content lifecycle.

• Actionable Alerts: Go beyond flagging issues — automatically trigger notifications to stakeholders within workflows, transforming compliance checks into immediate corrective actions.

Resources

• Markup AI Website
• n8n Community Nodes Documentation
• n8n Documentation
• GitHub Repository

Development

Contributions are welcome! This section is for developers who want to contribute to this community node or set up a local development environment.

Prerequisites

• Node.js >= 22.16
• npm
• Basic familiarity with n8n and TypeScript

Quick Start (Recommended)

After cloning this repository:

npm install
npm start

That's it. npm start runs n8n-node dev (the official @n8n/node-cli toolchain), which:

• builds the node in watch mode (rebuilds automatically on file changes),
• starts n8n in a subprocess with this node pre-loaded,
• opens the editor at http://localhost:5678.

You do not need n8n installed globally. The CLI bundles a compatible n8n version for you.

The dev n8n stores its data in ~/.n8n-node-cli/.n8n/ (separate from any other n8n install you may have at ~/.n8n/).

First-time SQLite fix: If npm start errors with SQLite package has not been found installed, the sqlite3 native binding in the npx cache needs to be rebuilt for your Node version. The @n8n/node-cli toolchain caches n8n under a hash-named directory; this one-liner finds it and rebuilds:

cd "$(find ~/.npm/_npx -maxdepth 4 -type d -path '*/node_modules/n8n' -print -quit | xargs dirname | xargs dirname)"
npm rebuild sqlite3

Then re-run npm start.

Manual Setup (alternative)

If you prefer to run an n8n instance you already have installed (e.g. global install, Docker, an existing ~/.n8n/), use the npm link workflow:

1. Install dependencies after cloning this repository

npm install

2. Build the code

npm run build

3. Link the build from the dist folder

cd dist
npm link
cd ..

4. Create a custom directory inside n8n if it does not exist

mkdir -p ~/.n8n/custom
cd ~/.n8n/custom
npm init -y

5. Link the custom folder to the build

npm link @markupai/n8n-nodes-markupai

6. Fix n8n config file permissions (if needed)

chmod 600 ~/.n8n/config

7. Start your own n8n instance (e.g. n8n start if globally installed). You should now see Markup AI in the list of nodes.

npm run setup automates steps 1–6 of this flow.

Making Changes

1. Make your changes in the nodes/ or credentials/ directories
2. Build the code: npm run build
3. Test in n8n (refresh the browser if n8n is already running)
4. Run tests: npm test
5. Run linter: npm run lint:check

Testing

Run the test suite:

npm test

Run tests with coverage:

npm run test:coverage

Run tests in watch mode during development:

npm run test:watch

Troubleshooting

SQLite package has not been found installed on npm start: The sqlite3 native binding in the @n8n/node-cli npx cache wasn't built for your Node version. Rebuild it once:

cd "$(find ~/.npm/_npx -maxdepth 4 -type d -path '*/node_modules/n8n' -print -quit | xargs dirname | xargs dirname)"
npm rebuild sqlite3

Resetting the dev n8n instance: If you want a clean slate (forget owner setup, credentials, workflows), nuke the dev user folder:

rm -rf ~/.n8n-node-cli/

The next npm start will prompt you to create the owner account again.

Module loading errors (manual setup only): If you're using the Manual Setup path and see errors like require(...).index is not a constructor, make sure you've linked from the dist folder, not the project root. To fix:

npm run cleanup  # Remove old links
npm run setup    # Set up correctly

Permissions errors (manual setup only): If you see warnings about file permissions, run chmod 600 ~/.n8n/config to fix them.

Debugging setup issues: If you encounter problems during setup, enable debug mode for detailed output:

DEBUG=1 npm run setup    # Verbose setup with debug information
DEBUG=1 npm run cleanup  # Verbose cleanup with debug information

This will show:

• Exact commands being executed
• Full command output (both stdout and stderr)
• Detailed error information when commands fail
• Working directory for each command

License

MIT

Support

• For general n8n questions: n8n Community Forum
• For Markup AI API questions: Markup AI Support

Made with ❤️ by the Markup AI team