📘 Khy-OS 文档站 🧭 新手先读:核心概念

cli-anything-ollama


name: "cli-anything-ollama"
description: >-
Command-line interface for Ollama - Local LLM inference and model management via Ollama REST API. Designed for AI agents and power users who need to manage models, generate text, chat, and create embeddings without a GUI.


cli-anything-ollama#

Local LLM inference and model management via the Ollama REST API. Designed for AI agents and power users who need to manage models, generate text, chat, and create embeddings without a GUI.

Installation#

This CLI is installed as part of the cli-anything-ollama package:

pip install cli-anything-ollama

Prerequisites:

  • Python 3.10+
  • Ollama must be installed and running (ollama serve)

Usage#

Basic Commands#

# Show help
cli-anything-ollama --help

# Start interactive REPL mode
cli-anything-ollama

# List available models
cli-anything-ollama model list

# Run with JSON output (for agent consumption)
cli-anything-ollama --json model list

REPL Mode#

When invoked without a subcommand, the CLI enters an interactive REPL session:

cli-anything-ollama
# Enter commands interactively with tab-completion and history

Command Groups#

Model#

Model management commands.

CommandDescription
listList locally available models
showShow model details (parameters, template, license)
pullDownload a model from the Ollama library
rmDelete a model from local storage
copyCopy a model to a new name
psList models currently loaded in memory

Generate#

Text generation and chat commands.

CommandDescription
textGenerate text from a prompt
chatSend a chat completion request

Embed#

Embedding generation commands.

CommandDescription
textGenerate embeddings for text

Server#

Server status and info commands.

CommandDescription
statusCheck if Ollama server is running
versionShow Ollama server version

Session#

Session state commands.

CommandDescription
statusShow current session state
historyShow chat history for current session

Examples#

List and Pull Models#

# List available models
cli-anything-ollama model list

# Pull a model
cli-anything-ollama model pull llama3.2

# Show model details
cli-anything-ollama model show llama3.2

Generate Text#

# Stream text (default)
cli-anything-ollama generate text --model llama3.2 --prompt "Explain quantum computing in one sentence"

# Non-streaming with JSON output (for agents)
cli-anything-ollama --json generate text --model llama3.2 --prompt "Hello" --no-stream

Chat#

# Single-turn chat
cli-anything-ollama generate chat --model llama3.2 --message "user:What is Python?"

# Multi-turn chat
cli-anything-ollama generate chat --model llama3.2 \
  --message "user:What is Python?" \
  --message "user:How does it compare to JavaScript?"

# Chat from JSON file
cli-anything-ollama generate chat --model llama3.2 --file messages.json

Embeddings#

cli-anything-ollama embed text --model nomic-embed-text --input "Hello world"
cli-anything-ollama embed text --model nomic-embed-text --input "Hello" --input "World"

Interactive REPL Session#

Start an interactive session for exploratory use.

cli-anything-ollama
# Enter commands interactively
# Use 'help' to see available commands

Connect to Remote Host#

cli-anything-ollama --host http://192.168.1.100:11434 model list

State Management#

The CLI maintains lightweight session state:

  • Current host URL: Configurable via --host
  • Chat history: Tracked for multi-turn conversations in REPL
  • Last used model: Shown in REPL prompt

Output Formats#

All commands support dual output modes:

  • Human-readable (default): Tables, colors, formatted text
  • Machine-readable (--json flag): Structured JSON for agent consumption
# Human output
cli-anything-ollama model list

# JSON output for agents
cli-anything-ollama --json model list

For AI Agents#

When using this CLI programmatically:

  1. Always use --json flag for parseable output
  2. Check return codes - 0 for success, non-zero for errors
  3. Parse stderr for error messages on failure
  4. Use --no-stream for generate/chat to get complete responses
  5. Verify Ollama is running with server status before other commands

More Information#

  • Full documentation: See README.md in the package
  • Test coverage: See TEST.md in the package
  • Methodology: See HARNESS.md in the cli-anything-plugin

Version#

1.0.1