{
  "id": "page:docs-hermes-research-raw-quickstart",
  "_kind": "Page",
  "_file": "wiki/docs/hermes-research/raw/quickstart.md",
  "_cluster": "wiki",
  "attributes": {
    "nodeKind": "Page",
    "sourcePath": "docs/hermes-research/raw/quickstart.md",
    "sourceKind": "repo-docs",
    "title": "Hermes Agent Quickstart",
    "displayName": "Hermes Agent Quickstart",
    "slug": "docs/hermes-research/raw/quickstart",
    "articlePath": "wiki/docs/hermes-research/raw/quickstart.md",
    "article": "\n# Hermes Agent Quickstart\n\n> Source: https://hermes-agent.nousresearch.com/docs/getting-started/quickstart\n\n## Overview\n\nHermes Agent is an AI agent that operates through a command-line interface, supporting multiple LLM providers and featuring tool access, messaging platform integration, and skill automation.\n\n## Installation\n\n**Desktop (macOS/Windows):** Download from https://hermes-agent.nousresearch.com/desktop\n\n**Command-line only:**\n- Linux/macOS/WSL2/Android: `curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash`\n- Windows PowerShell: `iex (irm https://hermes-agent.nousresearch.com/install.ps1)`\n\nAfter installation, reload your shell: `source ~/.bashrc` or `source ~/.zshrc`\n\n## Provider Selection\n\nRun `hermes model` for interactive setup. Key requirements: \"a model with at least **64,000 tokens** of context.\"\n\nMajor supported providers include:\n- Nous Portal (subscription, OAuth)\n- OpenAI/Anthropic/Google\n- OpenRouter, DeepSeek, xAI\n- Ollama, LM Studio (local)\n- AWS Bedrock, Azure Foundry\n- Hugging Face, and 30+ others\n\nQuick start: `hermes setup --portal` for Nous Portal with zero-config setup.\n\n**Configuration storage:**\n- Secrets -> `~/.hermes/.env`\n- Non-secret settings -> `~/.hermes/config.yaml`\n\n## First Chat\n\n```bash\nhermes              # Classic CLI\nhermes --tui        # Modern TUI interface\n```\n\nVerify: The agent responds with your chosen model and can use tools (terminal, file read, web search).\n\n## Core Commands\n\n| Command | Purpose |\n|---------|---------|\n| `hermes` | Start chatting |\n| `hermes --continue` | Resume last session |\n| `hermes model` | Switch providers/models |\n| `hermes doctor` | Diagnose issues |\n| `hermes gateway setup` | Connect messaging platforms |\n| `hermes skills browse` | Discover reusable workflows |\n\n## Key Features\n\n**Slash commands:** Type `/` for autocomplete (`/help`, `/tools`, `/model`, `/personality`)\n\n**Terminal access:** Agent can execute commands and show results\n\n**Multi-line input:** Alt+Enter, Ctrl+J, or Shift+Enter\n\n**Interruption:** New message during task execution switches focus; Ctrl+C also works\n\n## Advanced Setup (After Base Chat Works)\n\n- **Messaging bots:** `hermes gateway setup` for Telegram, Discord, Slack, WhatsApp, Signal, Email\n- **Sandboxing:** `hermes config set terminal.backend docker`\n- **Voice:** Install extras, then `/voice on` with Ctrl+B to record\n- **Skills:** `hermes skills install openai/skills/k8s` then `/k8s [task]`\n- **MCP servers:** Configure via `~/.hermes/config.yaml`\n\n## Troubleshooting\n\nCommon issues:\n- Empty replies -> Run `hermes model` to verify provider auth\n- Session won't resume -> Check `hermes sessions list`\n- Gateway unresponsive -> Re-run `hermes gateway setup`\n\nRecovery sequence: `hermes doctor` -> `hermes model` -> `hermes setup` -> `hermes sessions list`\n",
    "documents": []
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "page:docs-hermes-research",
      "to": "page:docs-hermes-research-raw-quickstart",
      "kind": "contains_page"
    }
  ]
}