Terminal
Terminal/Skill
Guides building semantic and hybrid search on Chroma Cloud, covering managed collections, cloud clients, schemas and the Chroma CLI.
By Chroma
Works with
npx skills add chroma-core/agent-skills --skill chroma-cloudDo not block on a long questionnaire. Ask only for details that are missing and required to choose the right path:
CHROMA_API_KEY, CHROMA_TENANT, and CHROMA_DATABASE are already configuredIf the user has no embedding preference, default to Chroma Cloud Qwen. If hybrid search is required, use Schema() and Search(). If the task is narrow, such as fixing an existing query, reviewing code, or answering an API question, proceed with the repo context instead of forcing intake.
CloudClient vs Client)Schema() and Search() are only used for Cloud workflowsget_or_create_collection() accepts either an embedding_function OR a schema, but not both. Use schema when you need multiple indexes, hybrid search, or sparse embeddings; use embedding_function for simple dense-only search.Use the CLI topic to authenticate and write Cloud credentials:
chroma login
chroma db create <db_name>
chroma db connect <db_name> --env-file
Then create a CloudClient and choose the API based on the search mode:
import { CloudClient } from 'chromadb';
const client = new CloudClient();
const collection = await client.getOrCreateCollection({ name: 'my_collection' });
Use collection.query() for dense-only search. Use Schema() plus Search() only when the user needs hybrid retrieval, multiple indexes, or more expressive ranking/query composition.
Collections are the main isolation boundary in Chroma Cloud, and metadata is the main filtering mechanism inside a collection. Reach for Schema() only when you need explicit dense+sparse or multi-index configuration, and reach for Search() only when query() is not expressive enough.
If you need more detailed information about Chroma beyond what's covered in this skill, fetch Chroma's llms.txt for comprehensive documentation: https://docs.trychroma.com/llms.txt