data
Chroma CLI for Shell Scripts
Use the Chroma CLI for shell scripts with headless JSON commands, schema discovery, credentials, and permission controls.
7 functions 5 read 2 write API key auth
Chroma CLI for Shell Scripts
Call integration functions from shell scripts with stable JSON input and output.
Use shell scripts for small local automations that need one or more integration calls. The Chroma CLI uses the same integration registry as the TUI, Lua runtime, and MCP gateway, but returns predictable command output for automation.
Command Shape
# Chroma CLI for Shell Scripts
kosmokrator integrations:configure chroma --set api_key="$CHROMA_API_KEY" --enable --read allow --write ask --json
kosmo integrations:call chroma.chroma_list_collections '{"limit":1,"after":"example_after"}' --json Discovery Before Execution
Agents and scripts can inspect Chroma docs and schemas before choosing a function.
kosmo integrations:docs chroma --json
kosmo integrations:docs chroma.chroma_list_collections --json
kosmo integrations:schema chroma.chroma_list_collections --json
kosmo integrations:search "Chroma" --json
kosmo integrations:list --json Useful Chroma CLI Functions
| Function | Type | Parameters | Description |
|---|---|---|---|
chroma.chroma_list_collections | Read | limit, after | List all vector collections in Chroma. Returns collection names and IDs that can be used for further operations. |
chroma.chroma_get_collection | Read | collection_id | Get details of a specific Chroma collection by its name or UUID, including metadata and document count. |
chroma.chroma_create_collection | Write | name, description, metadata | Create a new vector collection in Chroma. Collections are used to store and query document embeddings. |
chroma.chroma_add_documents | Write | collection_id, ids, embeddings, documents, metadatas | Add documents with embeddings to a Chroma collection. Each document requires an ID and either embeddings or text content. |
chroma.chroma_query_documents | Read | collection_id, query_embeddings, query_texts, n_results, where, where_document, include | Search for similar documents in a Chroma collection using query embeddings or text. Returns the most similar documents ranked by distance. |
chroma.chroma_get_document | Read | collection_id, ids, where, where_document, include, limit, offset | Retrieve specific documents from a Chroma collection by their IDs. Returns the full documents including text, embeddings, and metadata. |
chroma.chroma_get_health | Read | none | Check the health status of the Chroma vector database server. Returns heartbeat and version information. |
Automation Notes
- Use
--jsonfor machine-readable output. - Keep credentials out of argv by using environment variables or stored KosmoKrator configuration.
- Configure read/write policy before unattended runs; use
--forceonly for trusted automation. - Use the MCP gateway instead when the agent needs dynamic tool discovery inside a conversation.