CLI Reference

Complete reference for all llmswap CLI commands.

Table of contents

  1. Core Commands
    1. llmswap ask
    2. llmswap chat
    3. llmswap generate
    4. llmswap review
    5. llmswap debug
    6. llmswap providers
  2. Analytics Commands
    1. llmswap compare
    2. llmswap usage
    3. llmswap costs
  3. Configuration Commands
    1. llmswap config
  4. Advanced Commands
    1. llmswap logs
  5. Global Options
  6. Environment Variables
  7. Exit Codes

Core Commands

llmswap ask

Quick one-off questions to AI.

llmswap ask "Your question here"

Options:

  • --provider PROVIDER - Specify provider (openai, anthropic, gemini, etc.)
  • --no-cache - Disable response caching
  • --quiet - Minimal output
  • --age N - Adapt language for age N
  • --audience TYPE - Target audience (developer, student, business, etc.)
  • --teach - Enable teaching mode
  • --explain - Add detailed explanations

Examples:

# Simple question
llmswap ask "What is Docker?"

# Explain to a specific audience
llmswap ask "Explain kubernetes" --audience "business owner"

# Age-appropriate explanation
llmswap ask "What is AI?" --age 10

# Use specific provider
llmswap ask "Generate a UUID" --provider openai

llmswap chat

Start an interactive chat session with conversation memory.

llmswap chat

Options:

  • --provider PROVIDER - Initial provider
  • --no-cache - Disable caching
  • --age N - Set age context for session
  • --audience TYPE - Set target audience

Chat Commands:

  • /help - Show available commands
  • /provider - Show current provider
  • /switch <provider> - Switch to different provider
  • /clear - Clear conversation history
  • /stats - Show session statistics
  • /age N - Update age context
  • /audience TYPE - Update audience
  • /quit or /exit - End chat session

Example Session:

$ llmswap chat
🤖 Starting chat with claude-3-5-sonnet
You: Hello, my name is Sarah
Assistant: Hello Sarah! Nice to meet you. How can I help you today?
You: /switch gemini
🔄 Switched to gemini
You: What's my name?
Assistant: Your name is Sarah.
You: /stats
📊 Session Statistics:
Messages: 4
Total tokens: 156
Session duration: 2m 15s
You: /quit
👋 Goodbye!

llmswap generate

Generate code or commands from natural language descriptions.

llmswap generate "description of what you want"

Options:

  • --language LANG - Target programming language
  • --execute - Execute generated bash commands (with confirmation)
  • --save FILENAME - Save output to file
  • --explain - Add explanations to generated code
  • --provider PROVIDER - Use specific provider

Examples:

# Generate bash command
llmswap generate "find files larger than 100MB"
# Output: find . -type f -size +100M

# Generate Python code
llmswap generate "async function to fetch JSON from API" --language python

# Generate and save
llmswap generate "nginx reverse proxy config" --save nginx.conf

# Generate and execute (asks for confirmation)
llmswap generate "create Python project structure" --execute

llmswap review

AI-powered code review for your files.

llmswap review <file>

Options:

  • --focus AREA - Focus area: bugs, security, style, performance, general
  • --language LANG - Override detected language
  • --provider PROVIDER - Use specific provider

Examples:

# General review
llmswap review app.py

# Security-focused review
llmswap review api.js --focus security

# Performance review
llmswap review database.py --focus performance

# Style review
llmswap review components/Header.jsx --focus style

llmswap debug

Analyze errors and get debugging help.

llmswap debug --error "error message"
llmswap debug --file error.log

Options:

  • --error MESSAGE - Error message to analyze
  • --file FILE - File containing error/logs
  • --context - Include system context
  • --language LANG - Programming language context

Examples:

# Debug an error message
llmswap debug --error "TypeError: Cannot read property 'map' of undefined"

# Analyze error log
llmswap debug --file npm-debug.log

# Debug with context
llmswap debug --error "Connection refused" --language python --context

llmswap providers

Show status of all configured providers.

llmswap providers

Output Example:

╭─────────────┬──────────┬──────────────────────┬─────────────────────╮
│ Provider    │ Status   │ Model                │ Issue               │
├─────────────┼──────────┼──────────────────────┼─────────────────────┤
│ ANTHROPIC   │ ✓ Ready  │ claude-3-5-sonnet    │                     │
│ OPENAI      │ ✓ Ready  │ gpt-4o               │                     │
│ GEMINI      │ ✗        │                      │ Missing API key     │
│ COHERE      │ ✓ Ready  │ command-r-plus       │                     │
│ PERPLEXITY  │ ✗        │                      │ Missing API key     │
│ WATSONX     │ ✗        │                      │ Missing project ID  │
│ GROQ        │ ✓ Ready  │ llama-3.3-70b        │                     │
│ OLLAMA      │ ✓ Ready  │ llama3.1             │                     │
╰─────────────┴──────────┴──────────────────────┴─────────────────────╯

Analytics Commands

llmswap compare

Compare costs across providers for a given token count.

llmswap compare --input-tokens <N> --output-tokens <N>

Options:

  • --input-tokens N - Number of input tokens
  • --output-tokens N - Number of output tokens
  • --monthly - Show monthly cost projection

Example:

$ llmswap compare --input-tokens 1000 --output-tokens 500

Provider Cost Comparison (1000 input, 500 output tokens):
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Provider    │ Cost      │ Savings vs Most Expensive
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Ollama      │ $0.0000   │ 100.0%
Groq        │ $0.0001   │ 99.5%
Gemini      │ $0.0019   │ 90.5%
Claude      │ $0.0150   │ 25.0%
GPT-4       │ $0.0200   │ 0.0%
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

llmswap usage

View usage statistics and analytics.

llmswap usage [--days N]

Options:

  • --days N - Show last N days (default: 7)
  • --provider PROVIDER - Filter by provider
  • --export FILE - Export to CSV/JSON

Example:

$ llmswap usage --days 30

Usage Statistics (Last 30 days):
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Provider    │ Queries │ Tokens    │ Cost
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
OpenAI      │ 142     │ 45,231    │ $2.15
Anthropic   │ 89      │ 31,452    │ $1.87
Gemini      │ 203     │ 67,891    │ $0.45
Ollama      │ 567     │ 234,567   │ $0.00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Total       │ 1,001   │ 379,141   │ $4.47
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

llmswap costs

Get cost analysis and optimization recommendations.

llmswap costs

Output includes:

  • Current month’s spending
  • Cost breakdown by provider
  • Optimization recommendations
  • Potential savings

Configuration Commands

llmswap config

Manage llmswap configuration.

llmswap config <action> [options]

Actions:

  • set KEY VALUE - Set configuration value
  • get KEY - Get configuration value
  • unset KEY - Remove configuration key
  • show [SECTION] - Display configuration
  • reset [SECTION] - Reset to defaults
  • export --file FILE - Export configuration
  • import --file FILE - Import configuration
  • validate - Validate configuration
  • doctor - Run diagnostics

Examples:

# Set default provider
llmswap config set provider.default anthropic

# Set model for specific provider
llmswap config set provider.models.openai gpt-4-turbo

# View all configuration
llmswap config show

# Reset provider settings
llmswap config reset provider

# Export configuration
llmswap config export --file my-config.yaml

Advanced Commands

llmswap logs

Analyze log files with AI assistance.

llmswap logs --analyze <logfile>

Options:

  • --analyze FILE - Log file to analyze
  • --since TIME - Filter logs since time
  • --level LEVEL - Filter by log level
  • --pattern PATTERN - Search pattern
  • --correlate - Find correlated events

Examples:

# Analyze application logs
llmswap logs --analyze app.log

# Find errors in last 2 hours
llmswap logs --analyze system.log --since "2h ago" --level error

# Correlate events
llmswap logs --analyze nginx.log --correlate

Global Options

These options work with all commands:

  • --provider PROVIDER - Override provider selection
  • --no-cache - Disable response caching
  • --quiet - Minimal output mode
  • --debug - Enable debug output
  • --version - Show version information
  • --help - Show help for any command

Environment Variables

Configure providers via environment variables:

# OpenAI
export OPENAI_API_KEY="sk-..."

# Anthropic
export ANTHROPIC_API_KEY="sk-ant-..."

# Google Gemini
export GEMINI_API_KEY="..."

# IBM Watson
export WATSONX_API_KEY="..."
export WATSONX_PROJECT_ID="..."

# Groq
export GROQ_API_KEY="gsk_..."

# Cohere
export COHERE_API_KEY="..."

# Perplexity
export PERPLEXITY_API_KEY="pplx-..."

Exit Codes

  • 0 - Success
  • 1 - General error
  • 2 - Configuration error
  • 3 - Provider error
  • 4 - Network error
  • 5 - Authentication error

Need more help? Check out our tutorials for real-world examples.

Copyright © 2025 llmswap. Distributed under the MIT license.