{"repo": "adammpkins/llama-terminal-completion", "files": {"WEBSITE.md": {"7b43fd4eda36b472cc57b79c2442995615f1418f": {"t": 1785674474.016142, "n": 6073, "h": "b7bf8f1ad8ee4b92", "c": "# LlamaTerm Website Content\n\n## About Section\n\n### What is LlamaTerm?\n\nEver wish you could look up Linux commands or ask questions and receive responses from the terminal? You probably need a paid service, an API key with paid usage, or at least an internet connection, right? \n\n**Not with LlamaTerm.**\n\nWe run a Large Language Model (think ChatGPT) locally, on your personal machine, and generate responses from there. No cloud required. No API fees. Just you and your AI assistant.\n\n---\n\n## Documentation\n\n### Installation\n\n#### Quick Install (Recommended)\n\n```bash\ncurl -sSL https://raw.githubusercontent.com/adammpkins/llama-terminal-completion/main/install.sh | bash\n```\n\nThis downloads the latest release and installs it to `/usr/local/bin`.\n\n#### From Source\n\n```bash\ngit clone https://github.com/adammpkins/llama-terminal-completion.git\ncd llamaterm\nmake install\n```\n\nRequires Go 1.21+.\n\n---\n\n### Getting Started\n\nLlamaTerm works out of the box with [Ollama](https://ollama.ai) running locally.\n\n```bash\n# Install Ollama (if not installed)\ncurl -fsSL https://ollama.ai/install.sh | sh\n\n# Pull a model\nollama pull llama3.2\n\n# Start using LlamaTerm\nlt ask \"What is the difference between TCP and UDP?\"\n```\n\n---\n\n### Core Commands\n\n#### `lt ask <question>`\n\nAsk any question and get an AI response.\n\n```bash\nlt ask \"How do I find large files in Linux?\"\nlt ask \"Explain recursion in simple terms\"\n```\n\nOptions:\n- `-c, --copy` — Copy response to clipboard\n\n#### `lt cmd <description>`\n\nGenerate a shell command from a natural language description.\n\n```bash\nlt cmd \"find all .go files modified in the last week\"\nlt cmd \"compress all images in this folder\"\n```\n\nOptions:\n- `--dry-run` — Show command without running\n- `-y, --yes` — Run immediately without confirmation\n\nLlamaTerm detects dangerous commands (`rm -rf`, `sudo`, etc.) and warns you before execution.\n\n#### `lt quick <description>`\n\nGenerate and run a command immediately (no confirmation).\n\n```bash\nlt quick \"show disk usage\"\nlt quick \"list running docker containers\"\n```\n\n#### `lt chat`\n\nStart an interactive chat session with conversation memory.\n\n```bash\nlt chat\nlt chat --resume  # Resume a previous conversation\n```\n\nIn-chat commands:\n- `/help` — Show available commands\n- `/model` — Switch models\n- `/history` — Browse saved conversations\n- `/new` — Start a new conversation\n- `/clear` — Clear current chat\n- `Ctrl+H` — Open conversation history\n- `Ctrl+O` — Open model selector\n- `Ctrl+C` — Exit (auto-saves)\n\n#### `lt explain <file> [question]`\n\nAnalyze and explain code or file contents.\n\n```bash\nlt explain main.go\nlt explain config.yaml \"What does this configure?\"\nlt explain error.log \"What went wrong?\"\n```\n\n#### `lt fix <error>`\n\nGet help fixing errors.\n\n```bash\nlt fix \"Error: module not found\"\nnpm run build 2>&1 | lt fix\n```\n\n#### `lt copy <question>`\n\nAsk a question and copy the response to clipboard.\n\n```bash\nlt copy \"Write a git commit message for adding user auth\"\n```\n\n---\n\n### Piping Input\n\nLlamaTerm reads from stdin, letting you pipe content:\n\n```bash\ncat error.log | lt ask \"What's wrong here?\"\ngit diff | lt ask \"Summarize these changes\"\ndocker logs myapp | lt fix\n```\n\n---\n\n### Chat History\n\nConversations are automatically saved and can be resumed later.\n\n```bash\n# List saved conversations\nlt history list\n\n# Resume a conversation\nlt chat --resume\n\n# Delete a conversation\nlt history delete <id>\n\n# Clear all conversations\nlt history clear\n```\n\n---\n\n### Configuration\n\nLlamaTerm can be configured via config file, environment variables, or CLI flags.\n\n#### Config File\n\nLocation: `~/.config/lt/config.yaml`\n\n```yaml\nbase_url: http://localhost:11434/v1\nmodel: llama3.2\napi_key: \"\"\nstream: true\nmax_tokens: 2048\ntemperature: 0.7\n```\n\nCreate a config file interactively:\n```bash\nlt config init\n```\n\nView current config:\n```bash\nlt config show\n```\n\n#### Environment Variables\n\n```bash\nexport LT_BASE_URL=https://api.openai.com/v1\nexport LT_MODEL=gpt-4o-mini\nexport LT_API_KEY=sk-...\n```\n\nOr use standard OpenAI environment variable:\n```bash\nexport OPENAI_API_KEY=sk-...\n```\n\n#### CLI Flags\n\n```bash\nlt --base-url https://api.openai.com/v1 --model gpt-4o-mini ask \"Hello\"\n```\n\n| Flag | Description |\n|------|-------------|\n| `--base-url` | API base URL |\n| `--api-key` | API key |\n| `-m, --model` | Model to use |\n| `--no-stream` | Disable streaming output |\n| `--max-tokens` | Maximum tokens to generate |\n| `--temperature` | Temperature for generation |\n\n---\n\n### Supported Providers\n\nLlamaTerm works with any OpenAI-compatible API:\n\n| Provider | Base URL | API Key Required |\n|----------|----------|------------------|\n| **Ollama** | `http://localhost:11434/v1` | No |\n| **LM Studio** | `http://localhost:1234/v1` | No |\n| **llama.cpp** | `http://localhost:8080/v1` | No |\n| **OpenAI** | `https://api.openai.com/v1` | Yes |\n| **Azure OpenAI** | Your deployment URL | Yes |\n| **Anthropic (via proxy)** | Proxy URL | Yes |\n| **Together AI** | `https://api.together.xyz/v1` | Yes |\n| **Groq** | `https://api.groq.com/openai/v1` | Yes |\n\n---\n\n### Shell Completion\n\nEnable tab completion for commands and flags:\n\n```bash\n# Bash\nlt completion bash > /usr/local/etc/bash_completion.d/lt\n\n# Zsh (add to ~/.zshrc)\nsource <(lt completion zsh)\n\n# Fish\nlt completion fish > ~/.config/fish/completions/lt.fish\n```\n\n---\n\n### All Commands Reference\n\n| Command | Description |\n|---------|-------------|\n| `lt ask <question>` | Ask a question |\n| `lt cmd <description>` | Generate a shell command |\n| `lt quick <description>` | Generate and run immediately |\n| `lt copy <question>` | Ask and copy to clipboard |\n| `lt chat` | Interactive chat session |\n| `lt explain <file>` | Explain code or file |\n| `lt fix <error>` | Help fix an error |\n| `lt config show` | Show configuration |\n| `lt config init` | Create config file |\n| `lt history list` | List saved conversations |\n| `lt history delete <id>` | Delete a conversation |\n| `lt history clear` | Clear all conversations |\n| `lt completion <shell>` | Generate shell completion |\n| `lt version` | Show version info |\n| `lt help` | Show help |\n"}}}}