Amazon S3

Connect an Amazon S3 bucket to Agent Context. Agent Context reads structured files from your bucket and makes them queryable with SQL. Each file becomes a table.

Setup

1. Go to Agent Context in the dashboard
2. Click Add Source → Amazon S3
3. Enter your credentials and the S3 path to the file
4. Click Test Connection to verify, then Save

Connection Parameters

Parameter	Required	Description
Region	No	AWS region (e.g., us-east-1)
Endpoint	No	Custom S3-compatible endpoint (for MinIO, Cloudflare R2, etc.)
Access Key	No	AWS access key ID
Secret Key	No	AWS secret access key
S3 URI	Yes	Path to the file: bucket/path/to/file.parquet
File Format	No	Auto-detected from extension. Override: csv, tsv, parquet, json, jsonl

Alternatively, you can paste a connection string with all parameters.

Supported File Formats

Format	Extension	Notes
Parquet	.parquet	Columnar format. Best performance for large datasets.
CSV	.csv	Auto-detects delimiter, headers, and column types.
TSV	.tsv	Tab-separated values.
JSON	.json	JSON array format.
JSONL	.jsonl	Newline-delimited JSON. One record per line.

S3-Compatible Services

The S3 connector works with any S3-compatible storage. Set the Endpoint field to your provider’s URL:

Service	Endpoint
Amazon S3	Leave blank (default)
MinIO	Your MinIO URL
Cloudflare R2	Your R2 endpoint
DigitalOcean Spaces	https://<region>.digitaloceanspaces.com
Backblaze B2	Your B2 S3-compatible URL

Troubleshooting

Issue	Fix
Access denied	Verify the access key has s3:GetObject and s3:ListBucket permissions on the bucket.
No tables found	Check that the S3 URI points to a supported file format.
Schema mismatch	Ensure the file has consistent column structure.
Timeout	Check network connectivity to the S3 region.