# rlaope/oh-my-hermes

Source: [GitHub](https://github.com/rlaope/oh-my-hermes)  
Feed7 permalink: https://feed7.dev/p/oh-my-hermes-0slra82  
Published: Unknown  
Trust: Needs Review (needs_review)

## Why Included

Oh My Hermes adds routing, parallel worktrees, verification gates and reviewed memory above Hermes. Its broad scope warrants testing the pinned installer and claims in isolation.

## Source Summary

Oh My Hermes packages model routing, coding delegation, governed workflows and reviewed long-term memory into one Hermes plugin. Its catalog claims **108 specialist skills**, while parallel work returns one of **four typed states** covering execution, schema, verification and integration readiness.

## Practical Implication

Treat it as a harness design reference before adopting the whole stack: visible routing, evidence-gated completion, file-isolated worktrees and memory provenance are reusable patterns. If installing, resolve and use a **full commit SHA** as its agent-install protocol specifies.

## Agent-Ready Context

Oh My Hermes packages model routing, coding delegation, governed workflows and reviewed long-term memory into one Hermes plugin. Its catalog claims **108 specialist skills**, while parallel work returns one of **four typed states** covering execution, schema, verification and integration readiness.

Treat it as a harness design reference before adopting the whole stack: visible routing, evidence-gated completion, file-isolated worktrees and memory provenance are reusable patterns. If installing, resolve and use a **full commit SHA** as its agent-install protocol specifies.

The repository’s performance and cost comparisons are project-authored, and the supplied material does not describe an independent evaluation. The large skill catalog, installer and model-routing configuration also create a substantial review surface.

## Connected Context

Feed7 judgment across 778 accumulated Signals:

Oh My Hermes consolidates several previously separate harness patterns—routing, isolated parallel work, evidence-gated completion, and reviewed memory—into one plugin. That integration makes it a useful reference, but its large catalog and installer broaden the audit surface, reinforcing selective adoption and pinned provenance rather than treating feature count or project-authored comparisons as validation.

- [huangruiteng/loopx](https://feed7.dev/p/loopx-0j0o7ux) — LoopX independently reinforces durable state, ownership, and evidence gates as harness responsibilities; Oh My Hermes packages similar controls with routing and coding delegation.
- [What Does Done Even Mean? Agents and Paperclip's Liveness Model - Dotta, Paperclip](https://feed7.dev/p/what-does-done-even-mean-agents-and-paperclip-s-liveness-model-dotta-pap-0lx8wfc) — Paperclip’s evidence-based liveness model provides the conceptual basis for Oh My Hermes’s typed execution, verification, and integration states.
- [affaan-m/ECC](https://feed7.dev/p/ecc-0438yju) — Both offer broad, cross-cutting harness catalogs, so ECC’s warning about selective installation, verification, and pruning directly applies to Oh My Hermes’s review surface.
- [Harness Engineering: Building the Production Cage for Powerful Domain Agents — Mike Chambers, AWS](https://feed7.dev/p/harness-engineering-building-the-production-cage-for-powerful-domain-age-0la85zn) — The production-cage framing confirms that Oh My Hermes’s routing, memory, isolation, and workflow controls belong to the harness around the model.

## Context Map

- Layer: agent
- Domains: coding
- Topics: harness-engineering, multi-agent, agent-memory

## Uncertainty

- The repository’s performance and cost comparisons are project-authored, and the supplied material does not describe an independent evaluation. The large skill catalog, installer and model-routing configuration also create a substantial review surface.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.