llm
Access and interact with Large Language Models from the command line using Simon Willison's llm CLI tool. Supports OpenAI, Anthropic, Gemini, Llama, and dozens of other models via plugins. Features include chat sessions, embeddings, structured data extraction with schemas, prompt templates, conversation logging, and tool use. This skill is triggered when the user says things like "run a prompt with llm", "use the llm command", "call an LLM from the command line", "set up llm API keys", "install llm plugins", "create embeddings", or "extract structured data from text".
When & Why to Use This Skill
The LLM CLI tool skill empowers users to interact with a vast ecosystem of Large Language Models—including OpenAI, Anthropic, Gemini, and local models—directly from the command line. It streamlines AI workflows by providing a unified interface for chat sessions, prompt templating, structured data extraction, and embedding management, making it an essential utility for developers and power users seeking to integrate advanced AI capabilities into their terminal environment.
Use Cases
- Multi-Model Workflow Integration: Seamlessly switch between different LLM providers like Anthropic, OpenAI, and Google Gemini using a single, consistent CLI syntax.
- Automated Code & File Analysis: Pipe source code or documentation directly into the tool to perform instant bug reviews, summarizations, or explanations without leaving the terminal.
- Structured Data Extraction: Utilize predefined schemas to extract specific data points from unstructured text or images, ideal for automating data entry or processing tasks.
- Prompt Template Management: Create and manage reusable prompt templates to standardize complex AI tasks and ensure consistent outputs across different projects.
- Semantic Search & Embeddings: Generate and store embeddings for local files to build custom semantic search engines or retrieval-augmented generation (RAG) systems.
- Local Model Execution: Install plugins to run and interact with local models like Llama, providing a private and offline-capable AI development environment.
| name | llm |
|---|---|
| description | Access and interact with Large Language Models from the command line using Simon Willison's llm CLI tool. Supports OpenAI, Anthropic, Gemini, Llama, and dozens of other models via plugins. Features include chat sessions, embeddings, structured data extraction with schemas, prompt templates, conversation logging, and tool use. This skill is triggered when the user says things like "run a prompt with llm", "use the llm command", "call an LLM from the command line", "set up llm API keys", "install llm plugins", "create embeddings", or "extract structured data from text". |
LLM CLI Tool Skill
A CLI tool and Python library for interacting with Large Language Models including OpenAI, Anthropic's Claude, Google's Gemini, Meta's Llama, and dozens of others via remote APIs or locally installed models.
When to Use This Skill
Use this skill when:
- Running prompts against LLMs from the command line
- Managing conversations and chat sessions
- Working with embeddings for semantic search
- Extracting structured data using schemas
- Installing and configuring LLM plugins
- Managing API keys for various providers
- Using templates for reusable prompts
- Logging and analyzing LLM interactions
Quick Reference
Basic Commands
# Run a prompt
llm "Your prompt here"
# Use a specific model
llm -m claude-4-opus "Your prompt"
# Chat mode
llm chat -m gpt-4.1
# With attachments (images, audio, video)
llm "describe this" -a image.jpg
# Pipe content
cat file.py | llm -s "Explain this code"
Key Management
llm keys set openai
llm keys set anthropic
llm keys set gemini
Plugin Management
llm install llm-anthropic
llm install llm-gemini
llm install llm-ollama
llm plugins
Documentation Index
Core Documentation
- README.md - Project overview and quick start guide
- docs/setup.md - Installation and initial configuration
- docs/usage.md - Comprehensive CLI usage guide (prompts, chat, attachments, conversations)
- docs/help.md - Complete command reference and help text
Model Configuration
- docs/openai-models.md - OpenAI model configuration and features
- docs/other-models.md - Configuration for other model providers
Advanced Features
- docs/tools.md - Tool use and function calling with LLMs
- docs/schemas.md - Structured data extraction from text and images
- docs/templates.md - Creating and using prompt templates
- docs/fragments.md - Long context support using fragments
- docs/aliases.md - Creating model aliases
Embeddings
- docs/embeddings/index.md - Embeddings overview
- docs/embeddings/cli.md - Embeddings CLI commands
- docs/embeddings/python-api.md - Embeddings Python API
- docs/embeddings/storage.md - Embeddings storage system
- docs/embeddings/writing-plugins.md - Writing embedding plugins
Plugins
- docs/plugins/index.md - Plugin system overview
- docs/plugins/installing-plugins.md - Installing and managing plugins
- docs/plugins/directory.md - Plugin directory listing
- docs/plugins/tutorial-model-plugin.md - Tutorial: Creating a model plugin
- docs/plugins/advanced-model-plugins.md - Advanced plugin development
- docs/plugins/plugin-hooks.md - Plugin hooks reference
- docs/plugins/plugin-utilities.md - Plugin utility functions
Python API & Development
- docs/python-api.md - Python library API reference
- docs/logging.md - Logging system and SQLite storage
- docs/contributing.md - Contributing to LLM development
Reference
- docs/related-tools.md - Related tools and ecosystem
- docs/changelog.md - Version history and changes
Common Workflows
Starting a Conversation
# Start chat with context
llm chat -m gpt-4.1 -s "You are a helpful coding assistant"
# Continue a previous conversation
llm -c "Follow up question"
Working with Files
# Analyze code
cat script.py | llm "Review this code for bugs"
# Process multiple files
cat *.md | llm "Summarize these documents"
Structured Output
# Extract data with schema
llm -m gpt-4.1 "Extract person info" -a photo.jpg --schema name,age,occupation
Template Usage
# List templates
llm templates
# Use a template
llm -t summarize < article.txt