---
title: "hermes-dreaming"
description: "Nightly deterministic memory consolidation for Hermes: corroborate facts with real conversations, propose the few worth keeping, ask about stale memory, warn before memory fills up."
canonical: https://agentpluginsdirectory.com/plugins/hermes-dreaming
last-updated: 2026-10-05
---

# hermes-dreaming
Nightly deterministic memory consolidation for Hermes: corroborate facts with real conversations, propose the few worth keeping, ask about stale memory, warn before memory fills up.
- Slug: hermes-dreaming
- Publisher: Anton Vaskov
- Repository: https://github.com/itpartypattaya/hermes-dreaming
- Manifest: plugin.json
- Version: 2.1.1
- License: MIT
- Category (editorial): other
- Skills: 1 (dreaming)
- MCP servers: 0
- Stars: 4
- Repository created: 2026-08-16
- Repository last pushed: 2026-10-03
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/hermes-dreaming
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What hermes-dreaming does, in the publisher's words

Nightly, deterministic memory consolidation ("dreaming") for Hermes Agent, in the spirit of OpenClaw dreaming.

An agent's memory goes wrong in two quiet ways: useful facts never make it into durable memory, and durable memory fills up with stale or duplicated entries until the char limit silently stops all writes. Every night a Python script looks at what the agent knows and what people actually said, and wakes the agent only when there is something to decide. The script calls no LLM and never writes memory; the agent applies its proposals through the regular memory tool and reports in a few lines.

- Corroboration from real conversations. A fact counts only if humans mention it on distinct days. Cron prompts, compaction banners, URLs, untrusted chats and the agent's own threads never corroborate anything.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/itpartypattaya/hermes-dreaming/HEAD/README.md