Editorial review: Codex (AI-assisted source review), 2026-10-16.

Hermes Agent

Nous Research's open-source agent with persistent memory, reusable skills and messaging-platform access.

Nous Research · Agent frameworks and orchestration · framework

Published in Feed

Overview

The Hermes Agent repository describes a self-improving agent that creates reusable skills, searches previous conversations and maintains memory across sessions. It offers a terminal interface, a gateway for messaging platforms and scheduled automations. Model backends can be changed, including Nous Portal, OpenRouter, OpenAI and custom endpoints. The repository identifies the project as MIT-licensed.

This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.

Aymorphic analysis

Persistent memory creates value only when it stores the right context. Test correction and deletion of a remembered fact alongside task completion, especially when the same agent is reachable from several messaging accounts.

Limitations

Repository desk review only. Funding, star counts and historical desktop-release details were removed rather than treated as current evidence.

Key features

  • Persistent memory and skill creation
  • Messaging gateway
  • Scheduled tasks and model choice

Use cases

  • Recurring personal agent tasks
  • Assistants accessed through chat platforms

Pricing

MIT-licensed software; hosting and connected model usage incur separate costs.

Pricing reflects the source record and may change. Confirm current terms with the provider.

Dates and sources

Released
Not listed
Record updated
16 Oct 2026
Record last checked
16 Oct 2026

Explore the Feed

Other entries filed under agent frameworks and orchestration; inclusion does not imply equivalent features.

Browse the Feed