Skip to main content

AWS S3

Connect your Knowledge Repository directly to Amazon S3 to ingest documents automatically.

This connector allows you to point to a specific file or folder in a cloud bucket.

💡 Core Concepts​

Before configuring the connector, understand how the data flows.

1. Storage Sync vs. Indexing​

When you connect an S3 bucket, SVAHNAR does two things:

  1. Copies the targeted files from your S3 bucket into SVAHNAR's secure storage.
  2. Indexes these files so your AI agents can search and retrieve information from them during conversations.

2. Security & Access​

Documents imported via this connector become part of your Knowledge Repository.

  • Knowledge Repository Access: Agents with access to this Knowledge Base can read these files.

⚙️ Configuration Steps​

Follow these steps to securely link your S3 bucket.

Prepare AWS Credentials​

To access your bucket, SVAHNAR needs valid AWS credentials.

  1. Log in to your AWS Console.
  2. Create an IAM User with Read access (s3:GetObject, s3:ListBucket) to the target bucket.
  3. Generate an Access Key ID and Secret Access Key.
Use Keyvault

For security, do not hardcode your Secret Key. Store it in the SVAHNAR Key Vault and reference it in your configuration.

Identify Target Resource​

Decide exactly what you want to import:

  • Specific File: Points to a single document (e.g., documents/policy_2024.pdf).
  • Folder: Points to a prefix (e.g., backups/production/). All valid documents within this folder will be indexed.

Configure Connector​

Provide the following details in your configuration:

  • S3 Path: The file key or folder prefix.
  • Region: The AWS region where your bucket resides (e.g., us-east-1).
  • Credentials: Your Access Key and Secret Key.

💡 Tip: Large Archives​

This connector is ideal for:

  • Large Archives: Importing legacy document stores without manual uploading.
  • Regular Syncs: If your production system dumps logs or reports to S3, this connector keeps your AI agents up-to-date with the latest data. (Note: For Enterprise users, regular syncs are available as a paid feature.)

🚑 Troubleshooting​

  • Access Denied (403)

    • Verify the IAM user has s3:GetObject permissions for the specific files.
    • If syncing a folder, ensure s3:ListBucket is also allowed.

  • File Not Found

    • Check the s3_path. S3 paths are case-sensitive.
    • Ensure there is no leading slash (/) in the path (e.g., use myfolder/doc.pdf instead of /myfolder/doc.pdf).

  • Region Mismatch

    • If you see "Bucket not found" errors, verify that the region_name matches exactly where the bucket was created (e.g., us-west-2 vs us-east-1).

💰 Additional Credit Consumption​

The KnowledgeRepo tool consumes credits based on the number of pages in the documents it indexes. Each page (up to legal size) consumes one credit.