logo-darklogo-darklogo-darklogo-dark
  • Tool Categories
    • 🎨Art & Creative Design505
    • 🏢Business Management644
    • 💻Coding & Development514
    • 👮Detection83
    • 🧠General Use728
    • 🏥Health & Wellness55
    • 📷Image & Photo Analysis100
    • 🖼️Image Generation & Editing618
    • 📐Interior & Architectural Design37
    • 🎓Learning & Education483
    • ⚖️Legal & Finance90
    • 🎭Lifestyle & Entertainment236
    • 📢Marketing & Advertising627
    • 🎧Music & Audio138
    • 👔Office & Workplace1,014
    • 🔬Research & Data Analysis373
    • 👥Social Media245
    • 🎥Video Generation & Editing426
    • 👧🏻Virtual Companion135
    • 🎤Voice Generation & Editing381
    • ✍️Writing & Editing808
    • All Categories
    • AI Use Cases
  • News
  • Events
    • Academic Conferences
    • Developer Conferences
    • Expos / Trade Shows
    • Industry Summits
    • Workshops / Training
    • All Events
    • Past Events
  • Saved Tools
  • Suggest a Tool
✕
Home › News › Nous Research ships Bot Mode for Hermes Agent, enabling multi-agent collaboration out of the box

Nous Research ships Bot Mode for Hermes Agent, enabling multi-agent collaboration out of the box

August 18, 2026
Nous Research ships Bot Mode for Hermes Agent, enabling multi-agent collaboration out of the box

Nous Research just made multi-agent coordination a default feature rather than a weekend project. The company announced Bot Mode for its open-source Hermes Agent framework, shipping it as a default-on feature in Hermes Desktop v0.20.3. What started as a one-day public beta plugin built by co-founder Teknium has graduated into something far more practical for developers who want multiple specialized agents working together without stitching together their own orchestration layer.

The core idea is straightforward. Each bot is a named Hermes profile stored on disk at ~/.hermes/profiles/. Memory, credentials, chat history, skills, and configuration all live in that profile, isolated from every other bot. There’s no new storage layer, no extra database to manage. Users can pin a specific model to each bot, assign a custom persona via a SOUL.md file, set an avatar, enable specific skills, and configure routine schedules. So one bot could be a research assistant running a smaller, faster model while another handles code review with a heavier one.

Communication between bots runs through a persistent Agent Inbox. Handoffs happen via CLI using @mentions, where a user or bot tags another agent and triggers a response. Group chats support two to six bots, with up to three serial rounds per user message. If a specific bot is mentioned, it responds. If no one is tagged, all bots take turns and each can either reply briefly or pass. It’s a sensible design that avoids the chaos of every agent speaking at once.

This matters because multi-agent orchestration is one of the messier problems in applied AI right now. Frameworks like LangGraph, AutoGen, and CrewAI have approached it with varying degrees of complexity. Nous Research’s angle is different: keep everything file-based, local, and MIT-licensed, which makes it attractive for developers who don’t want cloud dependencies or proprietary tooling. The open-source community building on Hermes now has a native way to build agent teams without writing custom routing logic from scratch.

  • Each bot runs as an isolated Hermes profile with separate memory and credentials
  • Bots communicate via a persistent Agent Inbox using @mention-based handoffs
  • Group chats support two to six bots with up to three response rounds per message
  • Fully MIT-licensed and bundled by default in Hermes Desktop v0.20.3

For founders and developers evaluating agent infrastructure, Bot Mode is worth watching. It won’t replace purpose-built orchestration platforms for complex enterprise pipelines, but for local-first, open-source workflows it lowers the bar considerably. The fact that it’s on by default means adoption happens without any configuration tax, and that’s often what separates a feature people actually use from one they mean to try someday.

Share

Related news

Michael Burry calls out Big Tech’s $3 trillion AI spending problem hiding in plain sight
August 18, 2026

Michael Burry calls out Big Tech’s $3 trillion AI spending problem hiding in plain sight


Read more
OpenAI launches ChatGPT for Teens with automatic age detection and stricter safety limits
August 18, 2026

OpenAI launches ChatGPT for Teens with automatic age detection and stricter safety limits


Read more
Honor’s next wide foldable may run on a 2nm Snapdragon chip with a 7,000mAh battery
August 17, 2026

Honor’s next wide foldable may run on a 2nm Snapdragon chip with a 7,000mAh battery


Read more

Recent Posts

  • Nous Research ships Bot Mode for Hermes Agent, enabling multi-agent collaboration out of the box
  • Michael Burry calls out Big Tech’s $3 trillion AI spending problem hiding in plain sight
  • OpenAI launches ChatGPT for Teens with automatic age detection and stricter safety limits
  • Honor’s next wide foldable may run on a 2nm Snapdragon chip with a 7,000mAh battery
  • Anthropic’s annualized revenue hits $65B, putting it ahead of OpenAI in the race to go public
Best AI Tools

Discover the best AI tools for any use case

Explore
  • Tool Categories
  • AI Use Cases
  • AI Events
  • AI News
  • Saved Tools
Company
  • About Us
  • Contact Us
  • Media & Partnerships
  • Suggest a Tool
Legal
  • Privacy Policy
  • Terms of Service
Copyright © 2026 Best AI Tools 415 Mission Street, 37th Floor, San Francisco, CA 94105