Hermes Agent

Tool

Hermes Memory Viewer — Browse & Search Agent Memory

Memory

Visualize and search your Hermes memory across all sessions with full-text search, timeline views, and date filtering.

Quick answer

A Hermes memory viewer lets you read and full-text search what the agent has stored across sessions instead of trusting it blindly. Hermes keeps memory as files and an FTS5-indexed store under ~/.hermes, so a viewer surfaces MEMORY.md, daily notes, and entity records with timeline and date filtering.

Hermes memory is inspectable by design. A viewer turns the on-disk MEMORY.md, summaries, and FTS5 index into something you can browse, search, and audit when the agent recalls the wrong thing.

Features

  • FTS5 search
  • Timeline view
  • Export
  • Filter by date

Why this tool matters

Hermes stores memory under ~/.hermes — human-readable Markdown (MEMORY.md, USER.md, SOUL.md) plus an FTS5 full-text index for fast retrieval. A viewer reads that same state, so what you see is exactly what the agent retrieves.

The most common memory complaint in the community is not that Hermes forgets, but that it recalls or saves the wrong thing. Being able to search the store by keyword and date is how you confirm whether a fact was ever written, and whether the agent is pulling stale context.

Because everything is local files, a viewer needs no cloud account and leaks nothing off your machine. That matters for the privacy-first audience that self-hosts specifically to keep conversation history on their own disk.

Memory viewers are read-and-search layers; the agent still owns the write path. If retrieval is consistently off, inspect the store, then tune what the agent is told to save rather than editing the index by hand.

Best use cases

Confirm whether a fact was ever saved before blaming the model
Audit what personal data Hermes is keeping on disk
Search session history by keyword and date range
Debug 'it forgot' reports by checking the FTS5 store directly
Review memory growth over time on a long-running agent
View on GitHub

FAQ

Where does Hermes store the memory a viewer reads?

Under ~/.hermes — Markdown files like MEMORY.md plus an FTS5 full-text index. A viewer reads that same on-disk state, so it shows exactly what the agent retrieves.

Why does Hermes sometimes recall the wrong thing?

Retrieval quality depends on what the agent chose to save and how it searches. A viewer lets you confirm whether a fact exists in the store and whether stale context is being pulled, which is usually the real issue rather than 'forgetting'.

Does a memory viewer send my data anywhere?

No. Hermes memory is local files, so a viewer reads from disk and needs no cloud account. That keeps conversation history on your own machine.

Related Resources