hermes-studio-dashboard

# Hermes Studio Dashboard > Skill by [ara.so](https://ara.so) — Hermes Skills collection. Hermes Studio is a comprehensive web dashboard and desktop application for [Hermes Agent](https://github.com/NousResearch/hermes-agent). It provides a unified control plane for managing AI agent conversations, platform integrations (Telegram, Discord, Slack, WhatsApp, Matrix, Feishu, WeChat, WeCom), session management, scheduled automation, model configuration, and usage analytics. Built with TypeScript and Vue3, it runs as a self-hosted local runtime, npm CLI package, or Docker container. ## Installation ### Desktop App (Recommended) Download the native installer for Windows, macOS, or Linux from [GitHub Releases](https://github.com/EKKOLearnAI/hermes-web-ui/releases/latest). Hermes data location: - **Windows**: `%LOCALAPPDATA%\hermes` (fallback: `%APPDATA%\hermes`) - **macOS/Linux**: `~/.hermes` ### NPM CLI ```bash npm install -g hermes-web-ui # Start the server hermes-web-ui start # Start with custom port hermes-web-ui start --port 3001 # Start with custom Hermes home HERMES_HOME=/custom/path hermes-web-ui start ``` ### Docker ```bash docker pull ekkolearnai/hermes-studio:latest docker run -d \ --name hermes-studio \ -p 3000:3000 \ -v ~/.hermes:/root/.hermes \ -v ~/.hermes-web-ui:/root/.hermes-web-ui \ -e AUTH_TOKEN=your-secure-token \ ekkolearnai/hermes-studio:latest ``` Access at `http://localhost:3000` ## Core Architecture Hermes Studio runs a Node.js backend (`packages/server`) that: - Bridges to Hermes Agent Python runtime via HTTP and Socket.IO - Manages its own SQLite database for Web UI sessions, users, jobs, Kanban tasks - Reads Hermes `state.db` for historical agent sessions (read-only) - Writes platform credentials to `~/.hermes/.env` - Writes channel behavior config to `~/.hermes/config.yaml` The Vue3 frontend (`packages/client`) connects via REST API and Socket.IO for real-time chat streaming. ## Authentication ### Default Credentials First login after fresh install: - **Username**: `admin` - **Password**: `123456` **Important**: Change default credentials immediately after first login via Settings → Account Management. ### Token-Based Auth Set a custom auth token: ```bash export AUTH_TOKEN=my-secure-token-123 hermes-web-ui start ``` Or in Docker: ```bash docker run -d \ -e AUTH_TOKEN=my-secure-token-123 \ -p 3000:3000 \ ekkolearnai/hermes-studio:latest ``` ### CLI Maintenance ```bash # Clear login IP locks hermes-web-ui clear-login-locks # Clear locks and restart server hermes-web-ui clear-login-locks --restart # Reset admin account to admin/123456 hermes-web-ui reset-default-login ``` ## Profile Management Hermes Studio supports multiple isolated profiles, each with its own: - Configuration and credentials - Session history and cache - Models, providers, skills, plugins - Scheduled jobs and usage metrics - Uploaded files ### Creating Profiles ```typescript // Via API: POST /api/hermes/profiles const response = await fetch('http://localhost:3000/api/hermes/profiles', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${token}` }, body: JSON.stringify({ name: 'production-bot', description: 'Production customer support agent' }) }); ``` ### Switching Profiles ```typescript // Via API: POST /api/hermes/profiles/switch await fetch('http://localhost:3000/api/hermes/profiles/switch', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${token}` }, body: JSON.stringify({ profileName: 'production-bot' }) }); ``` ### Exporting/Importing Profiles ```bash # Export creates ~/.hermes/profiles/{profile-name}.tar.gz # Import via Web UI: Profiles → Import Profile → select .tar.gz file ``` ## Chat Integration ### Real-Time Chat via Socket.IO ```typescript import { io } from 'socket.io-client'; const socket = io('http://localhost:3000', { auth: { token: process.env.AUTH_TOKEN } }); // Start chat run socket.emit('chat-run', { sessionId: 'session-uuid', message: 'What is the weather in San Francisco?', profile: 'default', model: 'gpt-4' }); // Listen for streaming chunks socket.on('chat-chunk', (data) => { console.log('Chunk:', data.chunk); console.log('Tool calls:', data.toolCalls); }); // Listen for completion socket.on('chat-complete', (data) => { console.log('Final response:', data.response); console.log('Usage:', data.usage); }); // Listen for errors socket.on('chat-error', (error) => { console.error('Error:', error.message); }); ``` ### REST API Chat ```typescript // Create session const sessionResponse = await fetch('http://localhost:3000/api/hermes/sessions', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ name: 'Customer Support Chat', profile: 'default' }) }); const { sessionId } = await sessionResponse.json(); // Send message (non-streaming) const chatResponse = await fetch('http://localhost:3000/api/hermes/chat', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ sessionId, message: 'Help me debug this Python error', profile: 'default', model: 'claude-3-5-sonnet-20241022' }) }); const result = await chatResponse.json(); console.log(result.response); ``` ## Platform Channel Configuration ### Telegram Bot ```typescript // Configure via API: POST /api/hermes/channels/telegram await fetch('http://localhost:3000/api/hermes/channels/telegram', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ enabled: true, botToken: process.env.TELEGRAM_BOT_TOKEN, mentionControl: 'required', // 'required' | 'optional' | 'none' reactions: true, freeResponseChats: ['123456789'] // Chat IDs for always-respond }) }); ``` Writes to `~/.hermes/.env`: ```bash TELEGRAM_BOT_TOKEN=your-token-here ``` And `~/.hermes/config.yaml`: ```yaml telegram: enabled: true mention_control: required reactions: true free_response_chats: - 123456789 ``` ### Discord Bot ```typescript await fetch('http://localhost:3000/api/hermes/channels/discord', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ enabled: true, botToken: process.env.DISCORD_BOT_TOKEN, mentionControl: 'required', autoThread: true, reactions: true, allowedChannels: ['1234567890123456789'], ignoredChannels: [] }) }); ``` ### Slack Bot ```typescript await fetch('http://localhost:3000/api/hermes/channels/slack', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ enabled: true, botToken: process.env.SLACK_BOT_TOKEN, appToken: process.env.SLACK_APP_TOKEN, mentionControl: 'optional', handleBotMessages: false }) }); ``` ### WhatsApp ```typescript await fetch('http://localhost:3000/api/hermes/channels/whatsapp', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ enabled: true, mentionControl: 'none', mentionPattern: '@bot' }) }); ``` ### Matrix ```typescript await fetch('http://localhost:3000/api/hermes/channels/matrix', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ enabled: true, accessToken: process.env.MATRIX_ACCESS_TOKEN, homeserver: 'https://matrix.org', autoThread: true, dmMentionThreads: true }) }); ``` ## Model Management ### Add Custom Provider ```typescript // POST /api/hermes/providers await fetch('http://localhost:3000/api/hermes/providers', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ name: 'local-ollama', baseUrl: 'http://localhost:11434/v1', apiKey: 'ollama', // Ollama doesn't require real key type: 'custom' }) }); ``` ### Fetch Available Models ```typescript // GET /api/hermes/providers/{providerId}/models const response = await fetch( `http://localhost:3000/api/hermes/providers/${providerId}/models`, { headers: { 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` } } ); const models = await response.json(); // [{ id: 'llama3.2', name: 'Llama 3.2', ... }, ...] ``` ### Set Default Model ```typescript // PATCH /api/hermes/settings await fetch('http://localhost:3000/api/hermes/settings', { method: 'PATCH', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ defaultModel: 'gpt-4o', defaultProvider: 'openai' }) }); ``` ## Scheduled Jobs (Cron) ### Create Cron Job ```typescript // POST /api/hermes/jobs await fetch('http://localhost:3000/api/hermes/jobs', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': `Bearer ${process.env.AUTH_TOKEN}` }, body: JSON.stringify({ name: 'Daily Report', cronExpression: '0 9 * * *', // 9 AM daily profile: 'default', task: 'Generate and email daily analytics report', enabled: true }) }); ```