Hermes Ollama Usage Monitor: model tool for Hermes Agent

Hermes plugin showing live Ollama Cloud usage, plan versus API cost and threshold alerts

What Hermes Ollama Usage Monitor does

The plugin reads your authenticated Ollama settings page, which exposes per-model usage-bar shares, and falls back to the GET /api/usage endpoint with an API key. A statusbar chip shows a color-coded summary such as Pro S25% W35%, turning yellow at 75% and red at 90%. The desktop pane breaks out session and weekly usage, per-model request counts, API-equivalent cost, plan versus API price per million tokens, cache break-even rates, lifetime break-even and eight weeks of history.

Because Ollama's dashboard forgets everything when the week resets, the plugin keeps local history that survives resets. It can raise a macOS notification when weekly usage crosses 75% or 90% and post a one-line daily usage summary in chat. Data refreshes every 60 seconds, and failed requests mark retained readings as stale instead of showing zero. The session cookie can live in the macOS Keychain or in a file with restricted permissions. A standalone CLI also works without Hermes, and the plugin is MIT licensed.

Key features

Statusbar chip with color-coded session and weekly usage
Per-model request counts and usage-bar shares
Plan versus API cost comparison with cache break-even analysis
Eight-week local history that survives weekly resets
Threshold alerts at 75% and 90% weekly usage
Standalone CLI that works without Hermes

When to use it

Seeing how much of the weekly Ollama Cloud quota is left without opening the website
Checking whether a subscription or API pricing is cheaper at your real cache hit rate
Getting a warning before the weekly limit runs out

Who it is for: Hermes Agent users who run models on Ollama Cloud and want to track quota and cost.

How it fits with Hermes Agent

Built for Hermes Agent as a plugin with a Desktop statusbar chip and pane, and it also ships a CLI that runs without Hermes.

How to install Hermes Ollama Usage Monitor

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

hermes plugins install Kosello/hermes-ollama-usage-monitor

Requirements: An Ollama Cloud account with either a __Secure-session cookie (primary source) or an API key (fallback), and Hermes Desktop for the chip and pane

Note: The README marks it as a work in progress and warns to expect bugs, including parser breakage if Ollama changes its HTML.

FAQ

What is Hermes Ollama Usage Monitor?

Hermes Ollama Usage Monitor is a Hermes Agent plugin that tracks Ollama Cloud usage and shows quotas, per-model breakdowns and plan versus API economics in the Hermes Desktop statusbar and a detail pane.

How do I install Hermes Ollama Usage Monitor?

Run hermes plugins install Kosello/hermes-ollama-usage-monitor, enable it if prompted, set up the cookie or API key fallback, then fully quit and relaunch Hermes Desktop so the Python backend routes mount.

Does Hermes Ollama Usage Monitor need my Ollama session cookie?

The cookie is the primary source because the settings page includes the per-model usage shares needed for per-model estimates. Without it, the plugin uses GET /api/usage with an API key, which gives aggregate quota percentages and request counts but not per-model Ollama prices.