Mnemon Memory
PassAudited by ClawScan on May 10, 2026.
Overview
Mnemon is a coherent persistent-memory skill, but installing it adds disclosed hooks that can reuse stored memories in future OpenClaw conversations.
Install this only if you want persistent cross-session memory. Before setup, review the external mnemon project if possible; after setup, inspect the prompt and plugin settings, avoid storing secrets, and use the documented forget/gc/eject commands to control retained memories and hooks.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Users must trust the external mnemon binary and its update path, because the reviewed skill file only documents how to install and use it.
The skill relies on an externally installed executable, including an unpinned @latest Go install path, while no implementation code is included in the artifact set.
brew install mnemon-dev/tap/mnemon ... go install github.com/mnemon-dev/mnemon@latest
Install only from a source you trust; prefer a pinned or verified release if available, and review the upstream project before running setup.
After setup, the agent may receive memory-related guidance and reminders automatically across sessions.
The setup installs persistent OpenClaw hooks and plugin components that can influence agent behavior after installation.
Hook → `~/.openclaw/hooks/mnemon-prime/` (agent:bootstrap — injects behavioral guide); Plugin → `~/.openclaw/extensions/mnemon/` (remind, nudge, compact hooks)
Review `~/.mnemon/prompt/guide.md` and the OpenClaw plugin configuration, disable hooks you do not want, and use the documented eject command if you no longer want the integration.
Private or inaccurate memories could be reused later unless the user audits or deletes them.
Persistent memories can be recalled into later conversations, so incorrect, sensitive, or stale stored facts could influence future agent responses.
`remind` | on | Recall relevant memories + remind agent on each message
Follow the skill's warning not to store secrets, periodically review memory logs/status, and use forget or garbage-collection commands to remove unwanted memories.