480+ Blueprints, ready to build.

The first step is always the hardest. Search or browse to kick-start your next flow.

17 Cloud

45 Blueprints

Sort by:

Detect data anomalies in new S3 files with DuckDB SQL and email a CSV of the bad rows.

Trigger a Kestra flow automatically whenever new files land in an Amazon S3 bucket, then process and archive them.

Classify each Pub/Sub event with a Gemini model on Vertex AI, then escalate only the high-priority ones to BigQuery and Discord.

Run a controlled Cloudflare canary deployment with DNS cutover, cache purge, a human validation pause, and automatic rollback.

Trigger a Python data pipeline automatically whenever a new file lands in an Amazon S3 bucket.

Download a zip archive, unpack it, and fan out all extracted files to Amazon S3 in parallel.

Download a ZIP from an HTTP API every hour, unzip it, and upload the CSV to Amazon S3 with a timestamped key.

Scan an AWS DynamoDB table and pass the extracted rows to a downstream processing task.

Check BigQuery for fresh source rows and submit a Dataproc Serverless PySpark batch only when the data is ready.

Orchestrate the GCS to Dataproc to BigQuery lakehouse chain in one flow, with a serverless PySpark transform and a gold table load.

Sync AWS resource metadata into Postgres with CloudQuery, then flag public S3 buckets with SQL.

List objects in an Amazon S3 bucket by prefix and fan out over every file in parallel with Kestra.

Stream Google Cloud Pub/Sub messages into Postgres in real time, one governed Kestra execution per message, with safe upserts and Teams alerts.

After a scheduled BigQuery aggregation, publish a completion event to a Google Cloud Pub/Sub topic so downstream consumers and flows react.

Authenticate with Amazon ECR, pull a private container image, and run a Python script inside it.

Query Apache Druid, convert results to CSV, stage them on S3, and load into Amazon Redshift.

Trigger a Kestra flow automatically whenever a new message lands in an Amazon SQS queue.

Clone a Git DNS manifest and batch-apply records to each managed Cloudflare zone.

Submit a Spark job to Dataproc Serverless on a schedule, with no cluster to manage and cost that tracks the work.

Run a Python data script in a Docker container pulled from a private Google Artifact Registry repository.

Apply Terraform, then upsert the provisioned IP into Cloudflare DNS, purge the cache, and notify Slack.