term-llm

Terminal-first AI runtime for commands, chat, editing, tools, jobs, agents, and local workflows.

Docs hub: https://term-llm.com

Why it exists

• turn natural language into executable shell commands
• run persistent chat with tools and MCP servers
• edit files with model assistance
• support agents, skills, sessions, jobs, and local automation
• work with hosted or local models

$ term-llm exec "find all go files modified today"

> find . -name "*.go" -mtime 0   Uses find with name pattern
  fd -e go --changed-within 1d   Uses fd (faster alternative)
  find . -name "*.go" -newermt "today"   Alternative find syntax
  something else...

Install

curl -fsSL https://raw.githubusercontent.com/samsaffron/term-llm/main/install.sh | sh

Or:

go install github.com/samsaffron/term-llm@latest

30-second quickstart

No API key needed if you use Zen:

term-llm exec --provider zen "list files"
term-llm ask --provider zen "explain git rebase"
term-llm chat

If you already have a provider key:

export ANTHROPIC_API_KEY=your-key
# or OPENAI_API_KEY / GEMINI_API_KEY / OPENROUTER_API_KEY / XAI_API_KEY / NEARAI_API_KEY / SAMBANOVA_API_KEY

Read the docs

The detailed docs live at https://term-llm.com and are authored in Markdown in this repo, then built with Hugo.

• Getting started
• Guides
• Architecture
• Reference

Common entry points:

• Configuration
• Providers and models
• Web UI and API
• Search
• Usage
• Agents
• Skills
• MCP servers
• Memory
• Jobs
• Text embeddings
• Audio generation
• Music generation
• Usage tracking
• Transcription
• Notifications

License

MIT