zilliz-cli 0.1.0

CLI for Zilliz Cloud resource management and Milvus operations

Zilliz CLI

Command-line tool for Zilliz Cloud cluster management and Milvus vector database operations.

Requirements

• Python 3.10+

Installation

pip install -e .

For development:

pip install -e ".[dev]"

Quick Start

# 1. Configure API Key
zilliz configure
# or
export ZILLIZ_API_KEY=your-api-key

# 2. List clusters
zilliz cluster list

# 3. Set cluster context (auto-resolves endpoint)
zilliz context set --cluster-id in01-xxxx

# 4. Work with collections and vectors
zilliz collection list
zilliz vector search --collection my_col --data '[[0.1, 0.2, 0.3]]' --limit 5

Commands

Configuration

# Interactive setup
zilliz configure

# Set individual values
zilliz configure set api_key YOUR_KEY
zilliz configure set output json

# View values
zilliz configure get api_key
zilliz configure list

Context

# Set cluster context (endpoint auto-resolved)
zilliz context set --cluster-id in01-xxxx

# Set with specific database
zilliz context set --cluster-id in01-xxxx --database mydb

# Change database only
zilliz context set --database another_db

# Manual endpoint override
zilliz context set --cluster-id in01-xxxx --endpoint https://custom.endpoint.com

# View current context
zilliz context current

Cluster Management

# List all clusters
zilliz cluster list
zilliz cluster list --output json

# Describe a cluster
zilliz cluster describe --cluster-id in01-xxxx

# Create a serverless cluster
zilliz cluster create --name my-cluster --project-id p1 --region gcp-us-west1

# Delete a cluster
zilliz cluster delete --cluster-id in01-xxxx

Collection Operations

Requires a context to be set first (zilliz context set --cluster-id ...).

# List collections
zilliz collection list

# Describe a collection
zilliz collection describe --name my_collection

# Quick Setup: create with dimension and metric type
zilliz collection create --name my_collection --dimension 768 --metric-type COSINE

# Custom Setup: create with full JSON schema
zilliz collection create --body '{"collectionName": "my_col", "schema": {...}}'
zilliz collection create --body file://schema.json

Vector Operations

Requires a context to be set first (zilliz context set --cluster-id ...).

# Vector search
zilliz vector search --collection my_col --data '[[0.1, 0.2, 0.3]]' --limit 10
zilliz vector search --collection my_col --data '[[0.1, 0.2, 0.3]]' --filter "age > 20" --output-fields '["name", "age"]'

# Query by filter
zilliz vector query --collection my_col --filter "id in [1, 2, 3]" --limit 10
zilliz vector query --collection my_col --filter "age > 20" --output-fields '["name", "age"]'

# Insert data
zilliz vector insert --collection my_col --data '[{"id": 1, "vector": [0.1, 0.2], "name": "doc1"}]'

Version

zilliz version
zilliz version --output json

Global Options

Option	Description
--output, -o	Output format: json, table, text
--api-key	API key (overrides config file and env var)
--help	Show help

Configuration

Credential Resolution

API Key is resolved in order (first found wins):

1. --api-key CLI flag
2. ZILLIZ_API_KEY environment variable
3. ~/.zilliz/credentials file

Config Files

~/.zilliz/
  credentials     # API key
  config          # CLI settings + cluster context

Output Formats

# Table (default, human-friendly)
zilliz cluster list

# JSON (machine/agent-friendly)
zilliz cluster list --output json

# Plain text
zilliz cluster list --output text

Testing

# Run unit tests
pytest tests/unit/

# Run all tests (E2E tests skipped without API key)
pytest

# Run E2E tests (requires real API key)
ZILLIZ_API_KEY=your-key pytest -m e2e

# Run E2E tests with data plane (requires cluster)
ZILLIZ_API_KEY=your-key ZILLIZ_CLUSTER_ID=in01-xxxx pytest -m e2e

License

Apache-2.0