S3 Storage

S3-compatible storage operations — upload, download, list, delete, and presign objects in AWS S3, MinIO, Cloudflare R2, DigitalOcean Spaces, and other S3-compatible services.

Use S3 Storage when you need object storage in a workflow:

• Upload generated reports, exports, or user files to a bucket
• Download file content for processing or transformation
• List objects in a bucket (optionally filtered by prefix)
• Delete objects when cleaning up temporary files
• Generate time-limited presigned URLs for secure downloads

Connect an S3 Storage node to an AI Agent Tools handle to let the agent upload files, list bucket contents, or generate download links.

The result (accessed as {{ s3.* }}) shape:

{
  "success": true,
  "url": "https://my-bucket.s3.amazonaws.com/uploads/report.pdf",
  "content": null,
  "objects": [{ "key": "uploads/report.pdf", "size": 1024 }],
  "error": null
}

Each node run executes a single operation. Set Operation, then configure the fields shown for that operation.

Shown when operation is upload:

Shown when operation is download:

Shown when operation is list:

Shown when operation is delete:

Shown when operation is presign:

Configure on the Environment Variables page:

Upload a report after generating it in a Code node:

Manual Trigger → Code → S3 Storage (upload)

S3 Storage settings:

• Operation: upload
• Bucket: my-app-exports
• Object Key: reports/{{ input.date }}.json
• Content: {{ code.result }}

Access the result downstream:

{{ s3.success }}
{{ s3.url }}

Generate a presigned download link:

Webhook Trigger → S3 Storage (presign) → Return

S3 Storage settings:

• Operation: presign
• Bucket: my-app-uploads
• Object Key: {{ webhook.body.fileKey }}
• Expires In (seconds): 3600

Return the presigned URL with {{ s3.url }}.

import webhook from @tensorify/webhook-trigger:4.0.0
import s3 from @tensorify/s3:1.0.0

node trigger @tensorify/webhook-trigger:4.0.0 {
    path = "/upload"
    method = "POST"
}

node upload @tensorify/s3:1.0.0 {
    operation = "upload"
    bucket = "my-app-exports"
    key = "uploads/{{ webhook.body.filename }}"
    content = "{{ webhook.body.content }}"
}

trigger.payload -> upload.input

S3 Storage has a hidden On Error control output branch (error). Enable it via the "Error output handle" toggle at the bottom of the settings panel.

The error branch fires when the S3 API returns an error or the request fails at runtime. On the main path, failures are also reflected in {{ s3.success }} and {{ s3.error }} on the result output.

• S3-compatible endpoints: Set S3_ENDPOINT_URL for MinIO, R2, or other non-AWS providers — omit it for standard AWS S3.
• Bucket is always required: Unlike optional fields, bucket must be set for every operation.
• Download content: Use {{ s3.content }} after a download operation; it is empty for other operations.
• List results: Use {{ s3.objects }} after a list operation to iterate over matched keys.
• Check s3.success: The node completes on the main output even when the operation fails. Use an If node or the On Error branch to handle failures.
• Agent permissions: Restrict Allowed Operations (Agent) — exclude delete if the agent should not remove objects.

• Code — generate content to upload
• Return — return presigned URLs as HTTP responses
• AI Agent — connect S3 as an agent tool
• Environment Variables — store AWS credentials and endpoint URL