A local, fast, and private semantic memory layer that allows AI agents to store facts, decisions, and context with biological decay and relationship mapping.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install engram-memory
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install engram-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 Engram - Agent Memory?
Engram provides a robust, local-first solution for AI agents needing persistent memory without relying on external cloud APIs. It implements semantic search and a biological decay algorithm, ensuring that relevant memories remain salient while unused information fades over time. As one of the more sophisticated Openclaw Skills, it allows agents to maintain context across multiple sessions, recall past decisions, and store user preferences securely on the local machine.
By leveraging typed memories and relationship graphs, Engram transforms a simple text store into a structured knowledge base. It handles deduplication automatically and offers granular scoping, making it an essential tool for developers building long-term agentic workflows within the Openclaw Skills ecosystem.
Engram - Agent Memory Use Cases
- Recalling past architectural decisions in complex coding projects.
- Storing and retrieving user preferences across different agent sessions.
- Maintaining a searchable history of past conversations and context.
- Building a local knowledge graph of related facts, events, and relationships.
- Ensuring agents do not repeat mistakes by referencing past decision logs.
How Engram - Agent Memory Works
- The agent initializes a search to recall relevant context based on the current task or environment.
- New information, facts, or decisions are ingested into the local database using specific types and tags.
- The system automatically handles deduplication for entries with a similarity greater than 92%.
- Relationships are established between memory nodes to create a structured, interconnected graph.
- A biological decay algorithm manages memory salience, boosting items that are frequently recalled while allowing others to fade.
Engram - Agent Memory Setup
To begin using this memory layer among your Openclaw Skills, install the package via npm:
npm install -g engram-memory
Once installed, verify the installation and check the current state of your local memory store:
engram stats
Engram - Agent Memory Data Schema & Taxonomy
Engram organizes data using a typed system with specific metadata attributes to ensure high retrieval accuracy:
| Attribute | Description |
|---|---|
| Type | Categorization such as fact, decision, preference, event, or relationship |
| Scope | Defines visibility: global, agent, private, or shared |
| Salience | A numerical value representing memory strength, subject to biological decay |
| Tags | User-defined labels used for manual filtering and categorization |
| Relationships | Graph connections like supports, contradicts, caused_by, or supersedes |
Engram - Agent Memory Advanced Features
- Biological decay algorithm that automatically manages memory relevance and clears stale data.
- Built-in deduplication engine that auto-merges similar memories to prevent redundant storage.
- Relationship mapping functionality to link disparate facts and create logical event chains.
- Local-first architecture ensuring total privacy with no requirement for API keys or cloud connectivity.
- Granular scoping capabilities allowing for shared or private memory partitions between multiple agents.
SKILL.md
Loading
Related Openclaw Skills
Weather Intelligence Digest
An automated tool to generate comprehensive weather forecasts and alert summaries using real-time NOAA data.
Weather Intelligence Digest
A Python-powered automation tool that generates detailed weather briefings and alert monitoring reports using real-time NOAA/NWS data.
Nori Health Coach
An AI-powered health assistant that provides personalized coaching by analyzing data from popular wearables and manual logs.
Esri Workflow Smell Detector
A deterministic preflight scanner that identifies technical risks and workflow smells in ArcGIS Pro projects via pay-per-request automation.
Multi-Brain Protocol
A sophisticated reasoning protocol for Openclaw Skills that synthesizes perspectives from three different LLMs in parallel before the primary agent responds.
GitHub PAT Integration
A secure, token-based interface for managing GitHub repositories, pull requests, and issues without the complexity of OAuth.
Featured*
