Oh My Hermes

What is Oh My Hermes?

Oh My Hermes (OMH) is an open-source package that sits on top of Hermes Agent from Nous Research without patching or replacing it. You keep talking to Hermes, and OMH picks a skill and a model for each lane of work, then runs independent lanes in parallel. Each lane gets a work category such as ultrabrain, deep or quick, which sets its model chain and reasoning effort, with fallback to the next model if a provider rejects one. The dag-ulw workflow turns a goal into a graph of small nodes and has a separate node verify the result. It ships a hundred-plus expert skills, ten ulw workflows, its own terminal interface, and long-term project memory with written reasons for what is kept. Status labels distinguish planned, reported done and verified work. It is MIT licensed and installs with a shell script, Homebrew, npm or Bun.

Use cases

• Run big coding jobs as parallel agent graphs
• Route each task to a suitable model and effort
• Plan, build and verify changes in loops
• Keep long-term project memory across sessions
• Review code and attack builds with QA scenarios
• Delegate tasks to Claude Code or Codex
• Track cost per lane in a live terminal view

Who it's for

• Hermes Agent users
• Developers
• AI-agent power users
• Open-source contributors

Why use Oh My Hermes

• Matches work to a model and effort level per lane, so simple tasks avoid the most expensive chain.
• Parallel lanes get disjoint write scopes, which prevents agents from editing the same file at once.
• Status labels separate planned, reported-done and verified work, so a green result means a check passed.
• Reviewed long-term memory carries decisions into later sessions without silent writes.
• Leaves Hermes untouched, so it can be updated or removed with omh commands.

Oh My Hermes features

• Category-based model chains with reasoning effort per lane
• Automatic fallback when a provider rejects a model
• dag-ulw graph runs with a separate verification node
• Hundred-plus omh expert skills loaded on demand
• Ten ulw workflows including plan, work, loop, qa and perf
• Long-term memory with recall packs under a token budget
• OMH Native terminal interface with live lane tree
• Evidence labels such as Plan not run and Test verified
• User-written guardrail rules for tool calls
• ast-grep playbook and codegraph commands

Oh My Hermes integrations

• Hermes Agent
• Claude Code
• Codex
• Homebrew
• npm
• Bun
• Discord
• ast-grep

How to get started with Oh My Hermes

1. Install with the shell script, Homebrew, npm or Bun, then run omh setup.
2. Open Hermes as usual and describe your goal, or type a workflow trigger such as dag-ulw or ulw-plan.
3. OMH splits the goal into lanes, tags each with a category, and assigns the model chain for it.
4. Ready lanes run in parallel, a separate checker node tests the goal, and any failed nodes are repaired.
5. Review the report, which labels each stage by how sure OMH is that it happened.

Oh My Hermes limitations

• Requires Hermes Agent to be installed
• The product A/B benchmark has no measured run published yet
• Setup needs a model-chain configuration step
• Model costs come from your own providers

Oh My Hermes pricing

Oh My Hermes is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

Oh My Hermes alternatives

The main Oh My Hermes alternatives listed on Debuly are Claude Code, Aider, OpenHands, goose, Cline and Agent Swarm. The table compares pricing and platforms; order is A–Z.

1. 

   Agent Swarm

   Self-hosted, MIT-licensed system where a lead agent hands work to Docker-isolated Claude Code, Codex and other coding workers.

   Automation and agentsCodingDeveloper platforms
   Open sourceVisit Agent Swarm (official site)
2. Ai

   Aider

   Open-source AI pair programming tool that runs in your terminal and edits code in your local git repository using the LLM you choose.

   CodingDeveloper platforms
   Open sourceVisit Aider (official site)
3. 

   Claude Code

   Anthropic's coding agent that fixes bugs, writes tests and runs migrations from your terminal, IDE, desktop app, web or Slack.

   CodingAutomation and agentsDeveloper platforms
   PaidVisit Claude Code (official site)
4. 

   Cline

   Open-source AI coding agent for VS Code, JetBrains, the terminal and desktop that edits files and runs commands with your approval.

   CodingDeveloper platformsAutomation and agents
   Open sourceVisit Cline (official site)
5. 

   goose

   Open source AI agent that runs on your machine as a desktop app, CLI or API and connects to tools through MCP extensions.

   Automation and agentsCodingDeveloper platforms
   Open sourceVisit goose (official site)
6. 

   OpenHands

   Open-source AI coding agent platform with a free Agent Canvas, a Teams plan, and self-hosted Enterprise control

   CodingAutomation and agentsDeveloper platforms
   FreemiumVisit OpenHands (official site)

Oh My Hermes FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: none collected yet, so a monogram is shown.
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.