Oh My Hermes: rlaope's all-in-one layer for Hermes Agent
Overview
Developer rlaope introduced Oh My Hermes, an all-in-one layer for Hermes Agent, in an X post, and says it is available on GitHub.
The post describes more than 150 workflow skills, a Recall feature that carries over prior decisions and rejected options, and a terminal interface with inline diffs, rich tool calls and a live agent tree. Its dag-ulw command splits one goal into a graph of subagents, each lane running on its own model.
A later X post from the same developer, under the name oh-my-hermes, describes the project as a single-line install that bundles 150+ skills and a long-term memory layer. That post says the memory layer gives 2x retention and that the setup was tuned over 200B tokens of use. Neither post includes independent measurements of these claims.
Written by AI from the articles below · updated Oct 28, 9:49 AM ET
Check the sources:
Developments
2 developments
- Oct 28, 1:12 AM ET · 1 articleIndividual developer releases oh-my-hermes, a one-line installer packaging a Hermes agent setupX search: AI launch posts (introducing, just launched): oh-my-hermes packages a Hermes agent setup with 150+ skills and memory layer
- Oct 25, 1:21 PM ET · 1 articleOh My Hermes launched as an all-in-one layer for Hermes AgentX search: AI launch posts (introducing, just launched): Oh My Hermes launches as an all-in-one layer for Hermes Agent
Article timeline
The articles in this story. Times are ET.
HOPE.@rlaopeXoh-my-hermes packages a Hermes agent setup with 150+ skills and memory layerAIDeveloper rlaope releases oh-my-hermes, an open-source single-line install for the Hermes agent. The package bundles 150+ skills, a long-term memory layer the author says gives 2x retention, and multi-model subagent routing. The author says the setup was tuned over 200B tokens of use.

HOPE.@rlaopeXOh My Hermes launches as an all-in-one layer for Hermes AgentAIOh My Hermes is introduced as an all-in-one layer for Hermes Agent, with 150+ workflow skills, a Recall feature that carries over prior decisions and rejected options, and a TUI with inline diffs, rich tool calls, and a live agent tree. Its dag-ulw command turns one goal into a graph of subagents, each lane running on its own model. The project is available on GitHub.

Heat Trend
- Current comparable heat
- 8
- Peak comparable heat
- 8Oct 28, 10:00 AM ET
- Change in past 24 hours
- +178%
The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.