Hermes Studio is a self-hosted web UI and dashboard designed for the Hermes Agent, transforming it into a comprehensive workspace for managing AI agents. It targets power users running Hermes Agent locally, offering advanced control over multi-agent orchestration, task scheduling, memory visualization, and more, thereby streamlining complex agent workflows and enhancing productivity.

How It Works

Hermes Studio functions as a sophisticated frontend for Hermes Agent, connecting via an OpenAI-compatible backend. It utilizes real-time Server-Sent Events (SSE) for streaming agent output and tool calls. The architecture is built with React and TypeScript, featuring profile-scoped workspaces for agent isolation and robust backend integration for features like session persistence (with Redis fallback), cron job management, and multi-agent crew coordination. Its core innovation lies in the integrated cron job manager and advanced orchestration capabilities, differentiating it from simpler chat interfaces.

Quick Start & Requirements

• Primary install/run: Clone the repository, run pnpm install, then pnpm dev for local development. Docker Compose is also available for a combined Hermes Agent and Studio setup.
• Prerequisites: Node.js version 22.0.0 or higher. An OpenAI-compatible backend (e.g., Hermes Agent gateway, Ollama, LM Studio, vLLM). Redis is optional but recommended for session persistence.
• Links: Repository: https://github.com/JPeetz/Hermes-Studio, Changelog: CHANGELOG.md, Contributing: CONTRIBUTING.md, License: LICENSE.

Highlighted Details

• Cron Job Manager: Schedule, monitor, and control autonomous agent tasks directly from the UI without needing terminal access.
• Multi-Agent Orchestration: Create and manage named crews of specialized agents, dispatch tasks in parallel, and monitor live activity feeds with profile-scoped workspaces for isolation.
• Interactive Knowledge Graph: Visualize agent memory relationships using a force-directed graph with zoom, pan, and node interaction.
• Visual Workflow Builder: Design and execute Directed Acyclic Graph (DAG) structured task pipelines for agent crews.
• Cost Tracking: Monitor per-crew token usage and estimated API costs with a model-aware price table.

Maintenance & Community

The project is actively maintained, indicated by frequent commits and a detailed roadmap. It provides guidelines for contribution and security. No specific community channels (like Discord or Slack) are listed in the README.

Licensing & Compatibility

The project is released under the MIT License, permitting commercial use and closed-source linking. Full functionality, including sessions, memory, and jobs, requires integration with the Hermes Agent gateway; otherwise, it operates in a limited "Portable Mode."

Limitations & Caveats

A native Electron-based desktop application and a fully managed cloud/hosted version are planned features but are not yet implemented. The complete feature set is dependent on a Hermes Agent gateway backend; without it, the UI operates in a reduced "Portable Mode."