Skip to Content
ConnectorsOverview

Connectors Overview

Connectors are the bridge between Slim.io and your data infrastructure. Each connector establishes a secure, least-privilege connection to a specific service, enabling Slim.io to discover and scan data for sensitive information across cloud storage, databases, and SaaS platforms.

In the Customer Dashboard, this area is labeled Data Sources. “Data Source” and “Connector” refer to the same configured integration — a single object that owns credentials, scope, and connection method.

Connection Methods

Cloud storage providers (AWS S3, Google Cloud Storage, Azure Blob Storage) can be connected via two methods. Choose the method that fits your security and deployment posture.

MethodWhere It RunsWhen To Choose
AgentlessSlim.io’s managed control plane calls provider APIs directly using the credentials you grantFastest path to value. No infrastructure to deploy. Recommended for most cloud accounts.
ScannerA deployable scanner image runs inside your cloud environment (Docker, Kubernetes, Cloud Run) and reports findings back to Slim.ioRequired for BYOC (“bring your own cloud”) deployments where data must never leave your network boundary, or for high-volume environments that benefit from local compute.

Both methods use the same credentials, the same classifier library, and produce the same findings. The wizard prompts you to choose a method when you create a cloud connector. Database and SaaS connectors run in agentless mode.

The Customer Dashboard surfaces both options under the same Data Sources workspace — there is no separate “Agentless” menu.

Supported Providers

Cloud Storage

ProviderServiceAuth MethodStatus
Amazon Web ServicesS3IAM Role (cross-account assume role)GA
Google Cloud PlatformCloud StorageWorkload Identity Federation (WIF)GA
Microsoft AzureBlob StorageService Principal (OAuth2 client credentials)GA

Databases

ProviderAuth MethodStatus
PostgreSQLUsername / password or IAM authenticationGA
MySQLUsername / password or IAM authenticationGA
SnowflakeUsername / password or key-pair authenticationGA
SQL Server (MSSQL)SQL authentication or Active DirectoryGA
OracleUsername / passwordGA
IBM DB2Username / passwordGA
DatabricksPersonal access token or OAuthGA

SaaS & Collaboration

ProviderAuth MethodStatus
SlackBot token (OAuth2)GA
Microsoft TeamsService Principal (OAuth2 client credentials)GA
OneDriveService Principal (OAuth2 client credentials)GA
SharePointService Principal (OAuth2 client credentials)GA
Google DriveService account (domain-wide delegation)GA
SalesforceOAuth2 (connected app)GA

Authentication Methods

Each provider uses a different credential model, but all follow the principle of least privilege:

AWS — IAM Role

Slim.io assumes a cross-account IAM Role in your AWS account. You create the role with a trust policy that allows Slim.io’s AWS account to assume it. The role needs only s3:GetObject, s3:ListBucket, and s3:GetBucketLocation permissions.

GCP — Workload Identity Federation

Slim.io uses Workload Identity Federation (WIF) to authenticate without long-lived service account keys. You configure a WIF pool and provider in your GCP project, then grant the Slim.io service account access to your Cloud Storage buckets.

Azure — Service Principal

You create a Service Principal (app registration) in your Azure Active Directory and grant it Storage Blob Data Reader access on the target storage account. Slim.io authenticates using the client credentials flow (client ID + client secret).

Databases — Username / Password or IAM

Database connectors support standard username and password authentication. Where available, IAM-based authentication (AWS RDS IAM, GCP Cloud SQL IAM) is also supported. All connections are established with read-only permissions.

SaaS — OAuth2 or Bot Tokens

SaaS connectors authenticate via OAuth2 flows or platform-specific bot tokens. Slim.io requests only the minimum scopes needed to read content — no write access is ever requested.

All authentication methods are designed to avoid storing long-lived credentials where possible. AWS uses role assumption with temporary tokens, GCP uses WIF with short-lived tokens, database passwords are encrypted at rest, and OAuth tokens are automatically refreshed.

Connector Lifecycle

  1. Create — Configure the connector with provider credentials and target scope
  2. Test — Validate that credentials work and Slim.io can access the target resources
  3. Active — Connector is ready for scanning
  4. Scan — Trigger manual or scheduled scans against the connector
  5. Disconnect — Temporarily disable without deleting configuration
  6. Delete — Permanently remove the connector and associated scan history

Connector Scoping

Each connector can be scoped to control which data is scanned:

  • Bucket / Container Selection — Specify which buckets or containers to include (cloud storage)
  • Prefix Filters — Limit scanning to specific key prefixes or folder paths (cloud storage)
  • File Type Filters — Include or exclude file extensions (e.g., scan only .csv, .json, .parquet)
  • Schema / Table Selection — Choose which schemas and tables to include (databases)
  • Channel / Site Selection — Choose which channels, sites, or drives to include (SaaS)
  • Size Limits — Skip files above a configurable size threshold

Provider-Specific Guides

  • AWS S3 — IAM role setup, permission policies, and region configuration
  • Google Cloud Storage — WIF configuration and service account setup
  • Azure Blob Storage — Service Principal creation and storage account access
  • Databases — PostgreSQL, MySQL, Snowflake, SQL Server, Oracle, DB2, and Databricks setup
  • SaaS & Collaboration — Slack, Teams, OneDrive, SharePoint, Google Drive, and Salesforce setup
  • Cloud DLP Integration — Layer provider-native DLP services alongside Slim.io detection
Last updated on