Collections and records of Chroma and Chroma Cloud: metadata-filtered pages and nearest-neighbour search by embedding or record. Upserts and deletes wait for approval.
The screenshots show GateRook 2.0.0: the Chroma form, the endpoint selector, and this connector’s actual tool preview. The form and preview have not been saved. Select a screenshot to open it at full size.
1. Create a connection and enter the settings
In GateRook, open Connections, select Add connection, search for Chroma, choose its card and select Continue to configure.
Use a Chroma server or Chroma Cloud endpoint. A local persistent database folder is not a server URL.
Identify the tenant and database that contain your collections. A self-managed default setup commonly uses default_tenant and default_database; Chroma Cloud supplies its own values.
Enter a Name, for example chroma-demo. Your AI app uses this name to choose the connection.
Enter Server URL, Tenant and Database, then narrow Collections to the required collection names or patterns.
Under Authentication, choose the header scheme the server expects: Token in X-Chroma-Token (Chroma Cloud), Bearer token in Authorization, or No authentication for a server that intentionally checks none. Supply API key or token when required.
For a private CA, enter CA certificate (PEM) under Advanced.
Choose Development, Staging or Production to match the service. Review the connector’s limit and access settings before continuing.
2. Test the connection and add it to an endpoint
Select Continue to test. GateRook checks the settings from your computer. Review the result and resolve any connection, certificate or permission error before continuing.
If collections are missing, verify the tenant and database as well as the credentials. A query vector must have the dimensions used by the collection; inspect the collection before asking for similarity results.
After the test passes, select Continue to add to an endpoint. Choose the endpoint your AI app uses, review its tool modes, then select Save connection. GateRook offers default as the starter endpoint.
3. Use the connection from your AI IDE
After saving Chroma on the endpoint and passing its connection test, ask your AI app to use chroma-demo, or the name you chose. If its tools are missing, confirm the endpoint assignment and refresh the MCP connection or start a new chat.
The preview below shows this connector’s default tools before saving. The table explains what each tool can do; connection settings, credentials and endpoint policies can restrict access further.
Start with a small read request. Replace example resource names with existing resources your connection may access. Open Activity after the call to inspect its connection, tool, arguments, policy decision and result metadata.
Ask your AI, for example
- “List the collections available through chroma-demo.”
- “Describe the support_docs collection on chroma-demo.”
- “Using chroma-demo, fetch the document with ID faq-123 from support_docs.”
| Tool | What it does | Starts as |
|---|---|---|
vector_list_collections | List the permitted Chroma collections in the configured tenant and database. | Read only |
vector_describe_collection | Show a collection’s ID, tenant, database, available dimension and metric, collection metadata, embedding function name and record count. | Read only |
vector_get | Fetch records by text ID or read a bounded metadata-filter page, including stored documents. Include embeddings only when requested. | Read only |
vector_query | Find nearest records using an embedding vector or a stored record ID, with a metadata filter and topK limit. Text query input is not supported by this connector. | Read only |
vector_upsert | Writes up to 100 records to one collection; a record with an existing id replaces it. You approve the exact records. Never retried. | Approval |
vector_delete | Deletes records by id, up to 1,000 of one collection. You approve the exact ids; nothing else is deleted. Never retried. | Approval |
You can change each tool’s mode on the endpoint. Every call your AI makes appears in Activity.