AI Agent Platforms / All platforms / Hermes Agent

Hermes Agent

Self-improving personal agent with skills, memory, scheduling, and messaging channel access.

Hermes Agent is a personal AI agent from Nous Research that creates skills from experience, improves them during use, and keeps persistent knowledge across sessions. It is used by developers and individuals who run an agent on their own infrastructure and reach it from a terminal or messaging apps.

Documented features

• Accessible via CLI, Telegram, Discord, Slack, WhatsApp, Signal, and Email through a unified gateway
• Creates skills from experience and improves them during use
• Provides a terminal TUI with multiline editing, slash-command autocomplete, history, and streaming tool output
• Recalls past sessions using FTS5 search with LLM summarization and user modeling
• Runs scheduled automations with a built-in cron scheduler that delivers to any connected platform
• Spawns isolated subagents for parallel workstreams
• Supports multiple terminal backends including local, Docker, SSH, Singularity, Modal, Daytona, and Vercel Sandbox
• Works with Nous Portal, OpenRouter, OpenAI, and custom model endpoints with a model switch command

Feature summaries come from the project's own repository and documentation and describe the software as documented on the verification date, not an evaluation.