Neutron Agent Memory for Openclaw

A persistent memory solution for AI agents that leverages semantic search and structured context storage to maintain state across sessions.

naeemmaliki036
v1.0.1
Feb 6, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install neutron-agent-memory

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install neutron-agent-memory using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Neutron Agent Memory?

The Neutron Agent Memory skill provides a robust interface for storing and retrieving agent memory through the Neutron API. It allows developers to save textual information as seeds for semantic retrieval and manage structured agent contexts, ensuring that AI assistants can persist their knowledge and state between separate interactions. By utilizing this skill, you can build smarter agents within the Openclaw Skills ecosystem that recall past conversations and specific domain knowledge.

This skill is particularly effective for workflows requiring long-term memory, such as multi-session research, complex project management, or personalized user interactions. It bridges the gap between ephemeral chat sessions and durable, searchable knowledge bases.

Neutron Agent Memory Use Cases

  • Persisting agent context and personality across multiple user sessions.
  • Implementing semantic search over historical conversation threads and project documents.
  • Storing and retrieving episodic, procedural, or working memory for complex AI workflows.
  • Snapshotting full thread states to allow agents to recall the specific evolution of a discussion.

How Neutron Agent Memory Works

  1. The agent captures information or conversation threads and sends them to the Neutron API using the save or context-create operations.
  2. Data is stored in dual formats: structured Agent Context for metadata and Seeds for full-text semantic indexing.
  3. When context is needed, the agent performs a semantic search or fetches a specific context ID to retrieve relevant memory.
  4. The skill returns the most relevant historical snapshots or metadata to restore the agent's knowledge state.

Neutron Agent Memory Setup

Configure your environment variables to authenticate with the Neutron API:

export NEUTRON_API_KEY=your_key
export NEUTRON_APP_ID=your_app_id
# Optional: defaults to 1
export NEUTRON_EXTERNAL_USER_ID=1

Alternatively, you can store credentials in ~/.config/neutron/credentials.json with the following structure:

{
  "api_key": "your_key_here",
  "app_id": "your_app_id_here",
  "external_user_id": "1"
}

Verify the installation by running the test script provided in the Openclaw Skills package: ./scripts/neutron-memory.sh test.

Neutron Agent Memory Data Schema & Taxonomy

The skill manages data across two primary entities to ensure maximum searchability and structural integrity.

Entity Description Supported Formats
Seed Full thread snapshots indexed for semantic search. Text, Markdown, JSON, CSV, Email
Agent Context Truncated summaries and metadata for session tracking. JSON

Each thread snapshot includes a timestamp, the full post content, and all associated comments, ensuring that later seeds contain more context than earlier ones.

Neutron Agent Memory Advanced Features

  • Supports multiple specialized memory types including episodic, semantic, procedural, and working memory.
  • Implements interaction seeds that capture full thread snapshots for comprehensive semantic retrieval.
  • Advanced semantic search capabilities with configurable result limits and similarity thresholds.
  • Native support for multiple chat formats including claude_chat and gpt_chat types.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Github Stars: 0
forks: 0

Featured*