Connect Weaviate to Gateco

Weaviate is an AI-native vector database with built-in hybrid search (BM25 + vector), GraphQL, and module-based vectorization. Gateco connects to Weaviate as a Tier 2 connector, adding identity-aware retrieval and RAG access control to your Weaviate classes. Gateco's policy layer provides an alternative to Weaviate RBAC and native tenant isolation, enforcing vector database RBAC based on organizational identity without modifying your Weaviate schema.

Weaviate grep search uses the Like operator with wildcard patterns, not regular expressions. Requests with pattern_type: "regex" are rejected. Weaviate keyword search uses native BM25 and returns ranked results.

Step 1: Create a Weaviate cluster

If you already have a cluster or self-hosted instance, skip to Step 2.

Go to console.weaviate.cloud.
Click Create cluster.
Select Sandbox (free) or a paid tier.
Choose a region and click Create.
Wait ~2 minutes for the cluster to become available.

Step 2: Get your cluster URL and API key

Click into your cluster in the Weaviate Cloud console.
Copy the Cluster URL (e.g. https://my-cluster.weaviate.network).
Under API keys, click Create API key or copy an existing one.

Step 3: Add the connector in Gateco

Navigate to Connectors → Add connector → Weaviate.
Enter your cluster URL and API key.
Click Test connection.
Click Save.

Field	Example	Description
`url`	`https://my-cluster.weaviate.network`	`Weaviate cluster URL (no trailing slash)`
`api_key`	`weaviate-api-key-here`	`API key (leave blank for unauthenticated self-hosted)`
`class_name`	`Document`	`Weaviate class/collection name (PascalCase)`
`text_property`	`content`	`Property containing the document text for keyword/BM25 search`

Step 4: Configure the class

Ensure your Weaviate class has a text property and a vector property. If you are using Weaviate's own vectorizer modules (e.g. text2vec-openai), Gateco can query the stored vectors directly. If you manage embeddings externally, configure the class schema with vector dimensions to match.

Troubleshooting

Error	Cause	Fix
`Forbidden (403)`	`Wrong or missing API key`	`Verify the API key in Weaviate Cloud → cluster → API keys`
`Class not found`	`Wrong class name (case-sensitive)`	`Weaviate class names are PascalCase. Check exact spelling in the Weaviate console.`
`connection refused`	`Weaviate not reachable on port 8080`	`Verify URL and that port 8080 is open in your firewall`

Frequently asked questions

Does Gateco work with both Weaviate Cloud and self-hosted Weaviate?

Yes. For Weaviate Cloud, enter the cluster URL and admin API key from the console. For self-hosted Weaviate, enter your instance URL (e.g. http://localhost:8080). Self-hosted instances do not require an API key by default.

What API key should I use?

For initial setup, use an admin key (APIKEY type with full access). For production, consider creating a read-only key. Weaviate API keys are created in the Weaviate Cloud console under Cluster details → API keys, or configured in the authentication section of your self-hosted weaviate.yaml.

Does Gateco support Weaviate native multitenancy?

Not in the current release. Gateco connects to Weaviate classes without tenant partitioning. For multi-tenant workloads, Gateco's identity-aware retrieval policies provide an alternative to Weaviate native tenant isolation, gating which documents each user's AI agent receives based on organizational identity and group membership, without relying on Weaviate RBAC or tenant partitioning. If you do use Weaviate multitenancy, connect Gateco to one tenant at a time by specifying the tenant in search_config. Full native multitenancy support is on the Gateco roadmap.