Why I Built a New Memory Plugin for Hermes Agent

A developer built hermes-mempalace, a new memory plugin for Hermes Agent that stores full conversation transcripts as readable markdown files rather than opaque embeddings. The plugin integrates with the MemPalace CLI and provides tools for searching and reviewing past context, with profile isolation under the HERMES_HOME directory.

Hermes Agent https://github.com/NousResearch/hermes-agent already has memory, and that matters. It keeps local context, it improves over time, and it works without forcing you into a cloud service. It also supports several external memory providers. I still built hermes-mempalace https://github.com/abdallah/hermes-mempalace , because none of the existing options fit my setup quite right. I wanted something: That last part is the important one. I did not want a memory layer that turns conversations into an opaque pile of embeddings or summaries you cannot really audit. I wanted actual transcripts, mined into a readable structure, with no hidden server in the middle. Why the existing options were not enough Hermes already gives you a few paths: And to be clear, some of those options are good. But ... I run Hermes on a headless machine at home. And I use separate profiles for different contexts. And I do not want conversation content depending on a cloud API or a separate service unless there is a very good reason. So, the best fit had to check a few boxes: HERMES HOME That let to MemPalace https://github.com/MemPalace/mempalace , or https://mempalaceofficial.com/ https://mempalaceofficial.com/ hopefully, that's the right one hermes-mempalace wires MemPalace into the Hermes memory provider interface. It follows the same lifecycle as the rest of Hermes memory providers: system prompt block adds a short memory reminder to the prompt. prefetch can run a MemPalace search before the first model call. sync turn buffers completed turns without slowing the chat loop. on session end writes buffered turns to markdown and mines them into the palace. shutdown flushes anything still waiting.During a session, Hermes gets three tools: mempalace search mempalace wake up mempalace status That gives the agent a simple pattern: find relevant past context, surface it early, and keep the full record available if you need to inspect it. The plugin stores data under the current Hermes profile, not in some shared global bucket. $HERMES HOME/mempalace/palace/ palace index $HERMES HOME/mempalace/conversations/ mined transcripts as markdown That was important to mw. Profile isolation should be structural, not something you hope the config handles correctly. If you already use Hermes: hermes memory setup Install the MemPalace CLI: uv tool install mempalace Then verify everything is in place: hermes memory status hermes plugins list The main difference is not flashy. It is that mempalace wake up gives Hermes useful context at the start of a session without turning the whole memory layer into a black box. You can inspect the transcripts. You can search them. You can open the directory and see what happened. If you want to prune or review something, you are not fighting the system. That is the part I cared about most. Hermes still keeps its built-in memory. hermes-mempalace does not replace that. It adds an archival layer on top of it. One that feels a lot more like a real notebook than a hidden model feature. If this sounds useful... let me know : If you are already happy with Hermes memory, you probably do not need this. But if you want: then hermes-mempalace may be a good fit. The plugin is MIT licensed, and the repo is here: