Who's Listening?

From its catalog listing

Memory that knows who's listening. Each memory is Private or Shared. Private memories go into a request only when the machine running Hermes (the gateway), the app you chat in (the platform) and the model provider are all trusted. By default, the gateway you install it on is trusted unless it's a hosted Hermes such as Hermes Cloud; the trusted platforms are the terminal and Hermes Desktop, including its Webapp; and the trusted model routes are Venice's private models and models on your own machine or local network. Everything else gets Shared memories only. You can configure all of these options in the plugin's settings: trusting either more or fewer gateways, platforms or model providers.

Good to know before installing

memories stay in $HERMES_HOME/whos-listening/. Once a day it fetches Venice's public model list from api.venice.ai (nothing about you is sent). With a Venice key in Hermes, a duplicate check (on by default) sends each new memory, each memory arriving by sync and each memory in a /memories review scan, together with up to 20 others, Private ones included, to Venice's private Clef model (duplicate_judge off keeps it local). Shared memories sync to your GitHub repository only if you set that up.

From its catalog listing.

How to install

  1. Install it. Hermes asks for any API keys it needs, then asks whether to turn it on.
    hermes plugins install whos-listening
  2. Said no? Turn it on later:
    hermes plugins enable whos-listening
  3. Restart Hermes so it loads the plugin.

Some plugins also need a sign-in or settings: see Full setup & settings. Memory plugins ask whether to become your memory provider instead. Steps follow the official plugins guide (opens in a new tab).

Category
Memory & notes
Maintained by
Web3B-eth (Community)
Needs
No API keys declared
Adds hooks
pre_llm_call post_tool_call
Needs Hermes
0.23.0 or newer
Version
0.1.0
In the catalog
2026-11-12

Known issues (from its catalog listing)

  • Hermes runs plugin middleware only on main chat requests. Compression, the tool-limit summary, mixture-of-agents and side calls skip the guard; see README Limits.
  • The first turn after a gateway starts carries Shared memories only, and Venice models count as Shared until Venice's model list has loaded on a new gateway.
  • Subagents and scheduled jobs count as Shared, so they never see Private memories.
  • Auto-review needs the default in-process plugin mode; under plugins.isolation host, held saves wait for the user.
  • For /memories to work from the first message, also run `hermes plugins enable whos-listening` (for a memory provider, install --enable only sets memory.provider).
  • If a model calls a memory tool through tool_call, Hermes answers that it's not a known tool name; the plugin tells models to call its tools directly.
Using the Hermes desktop app? See where plugins live
Hermes desktop Capabilities page, Plugins tab filtered to Official plugins built by Nous Research, each with an add button.
The official plugin catalog built into the app.Hermes desktop v0.22.5 · Capabilities > Plugins · Windows · captured Nov 1, 2026

Full setup & settings ↗Source code ↗

More in memory & notes

All memory & notes plugins

New Hermes Agent updates every day on X. Follow @HermesWatcher

← All plugins