SuperCompress

Hermes Agent · Updated 2026-10-09

Hermes Agent Token usage MCP

Reduce Hermes Agent Token Usage: Compress Tool Results

Hermes Agent runs long, tool-heavy sessions, and every tool result it keeps in context is sent again on later turns. Here is where Hermes tokens go and how SuperCompress plugs into Hermes so the agent keeps only the lines that answer the request it is working on.

October 9, 2026 · Guide

In short: run npm install -g supercompress-proxy && supercompress setup. Setup detects Hermes and adds an MCP server, shell hooks, a transform plugin and instructions, and turns on Hermes' own compression settings, so bulky tool results are compressed against the current request.

Where Hermes Agent tokens go

Habits that cut Hermes Agent token usage for free

  1. Start fresh for unrelated work. A new session drops old tool output.
  2. Keep instruction files short. Long AGENTS.md and persona files are sent on every turn.
  3. Only enable the MCP servers you use. Each one adds tool definitions.
  4. Use Hermes' built-in compression. Conversation compaction and tool-result pruning keep the session from growing without limit.

These habits shrink what goes in. They don't help when the agent genuinely needs to run a test suite or read a large file and only a few lines of the output matter. That is the part SuperCompress handles.

How SuperCompress works

SuperCompress is hosted, extractive and query-aware. Your agent sends a large block of context (a log, a diff, a file read, an MCP result) together with the question it is working on. Our Neural Keep model scores the lines against that question, keeps the original lines that answer it, and drops the rest. It never rewrites or summarizes, so file paths, error messages, IDs and code come back exactly as they were. The question itself is never compressed.

Every compress call reports original_tokens and kept_tokens, so you can see what was dropped on your own work instead of trusting a headline number. The usage_summary MCP tool shows totals per agent for your account.

What supercompress setup adds to Hermes Agent

WhatWhereWhat it does
MCP server~/.hermes/config.yaml (mcp_servers.supercompress)Gives the agent the compress_context, connect_account and usage_summary tools.
Shell hooks~/.hermes/agent-hooks/supercompress/pre_llm_call and post_tool_call hooks compress new bulky context against the current request.
Plugin~/.hermes/plugins/supercompress/A transform plugin that applies compression inside Hermes.
Instructions~/.hermes/AGENTS.mdTells Hermes to compress bulky dumps with compress_context and never to compress the request itself.
Hermes settings~/.hermes/config.yamlTurns on Hermes' native conversation compaction and proactive pruning of large tool results.

Install

npm install -g supercompress-proxy && supercompress setup

Setup links your SuperCompress account in the browser, detects the coding agents on the machine and wires each one. Restart the agent so it loads the new MCP server. Then check what is wired up:

supercompress doctor

doctor prints a per-agent health check. supercompress uninstall removes only the entries SuperCompress added. Setup backs up each config file before it edits it.

When to skip it

FAQ

Does SuperCompress change how I sign in to Hermes Agent?

No. You keep your normal Hermes Agent login and models. SuperCompress adds an MCP server and instructions to Hermes Agent; it does not route Hermes Agent's model traffic through a proxy.

Does SuperCompress summarize what Hermes Agent reads?

No. It is extractive. The lines it keeps are copied from the original output, so file paths, error messages and code stay exactly as they were.

How do I undo the Hermes Agent setup?

Run supercompress uninstall. It removes only the entries SuperCompress added, so your other MCP servers, hooks and instructions stay as they were.

Get API key Coding agents guide All coding agents