Build an integration

Ready to monitor, ingest, and visualize something? Let’s get started. You’ll build an integration package that defines how data is collected, processed, and presented.

• Overview and prerequisites: Get familiar with Elastic Stack fundamentals, including data streams, ingest pipelines, and field mappings, and review what you need before starting.
• Spin up the Elastic Stack: Set up and launch your local Elastic Stack environment for development and testing.
• Create a new package: Scaffold a new integration package using the elastic-package CLI, establishing the required directory structure and manifest files.
• Add a data stream: Define a new data stream to specify how your integration will collect and organize time-series data.
• Define deployment modes: Configure how your integration can be deployed, such as agent-based or agentless modes.
• Edit ingest pipelines: Customize ingest pipelines to parse, enrich, and transform incoming data before it’s indexed.
• Edit field mappings: Specify and adjust field mappings to define the structure and types of your indexed data.
• Create and export dashboards: Build Kibana dashboards and visualizations to explore and analyze the ingested data.
• Enable automatic installation for content packages: Enable automatic installation for your content package if data with a matching dataset is ingested in Elasticsearch.
• Testing and validation: Run tests and validate your integration to ensure it works as expected and meets quality standards.
• Finishing touches: Finalize documentation, metadata, and package details to prepare your integration for release.
• Tips for building integrations: Explore best practices and helpful advice to streamline your integration development process.