Agents
Hermes
An AI agent framework members use to orchestrate tasks, delegate to models, and automate workflows.
Also written as Hermes Agent
- Messages
- 1,566
- Members using it
- 132
- First seen
- Mar 9, 2026
What it means
A local or server-based agent framework that lets you build, run, and chain AI agents with access to tools, APIs, and multiple models. Members use Hermes to coordinate data pipelines, manage multi-platform workflows, and delegate tasks to models like GPT and Claude.
How it's used here
Members connect Hermes to data sources (like Grok bots feeding platform data), use it alongside Claude Code for orchestration, and troubleshoot issues with OpenAI subs in the Hermes subreddit.
Seen in the wild
“I have 1 grokbot right now feeding all the data per-platform to a Hermes agent with GPT Sol. Just got them going today. Will see how it goes”
“@rmktg @csoares12 why do you have Claude code talking to your Hermes? And how? And why not just deal with fable only / use Claude code only”
“If anayone got trouble with hermes + openai sub today : https://www.reddit.com/r/hermesagent/comments/1toq3lu/codex_open_ai_help_megathread”
More in this category
- AgentAn LLM instance given a role, tools, and instructions to autonomously perform tasks.
- Claude CodeAnthropic's coding interface that uses Claude models to write and edit code with a planning system.
- CodexAn AI coding tool members use for building and iterating on projects.
- SkillA reusable capability or prompt template you give an AI agent to perform specific tasks.