A persistent memory framework for AI agents to remember facts, track entities, and learn from past experiences.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-memory-1-0-0
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 agent-memory-1-0-0 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 AgentMemory?
AgentMemory is a robust, persistent memory system designed specifically for the Openclaw Skills ecosystem. It allows AI agents to move beyond transient context windows by storing long-term facts, tracking complex entities, and recording valuable lessons learned from past successes or failures. This skill acts as an external brain for your agents, ensuring they retain knowledge across multiple sessions and interact with users more intelligently over time.
By integrating AgentMemory into your developer workflow, you enable your agents to build a cumulative knowledge base. Whether it is remembering a user's specific technical preferences or documenting the outcome of a complex coding task, this Openclaw Skills addition provides the necessary infrastructure for truly autonomous and context-aware AI behavior.
AgentMemory Use Cases
- Loading relevant historical context at the start of every new agent session.
- Automatically storing durable facts extracted from long-form agent-user conversations.
- Documenting technical failures and recording corrective insights to avoid repeating the same mistakes.
- Maintaining a registry of people, projects, and roles through specialized entity tracking.
How AgentMemory Works
- The AI agent initializes the AgentMemory skill to establish a connection to its local persistent database.
- Relevant facts and entity metadata are stored using tagged entries for rapid semantic-style retrieval.
- The learning module captures specific actions, contexts, and outcomes to build a library of procedural insights.
- During subsequent tasks or sessions, the agent queries the memory database to recall contextually relevant information and historical lessons.
AgentMemory Setup
To get started with this entry in the Openclaw Skills library, install the package via the command line:
clawdhub install agent-memory
Initialize the memory system in your Python environment to begin persisting agent data:
from src.memory import AgentMemory
mem = AgentMemory()
AgentMemory Data Schema & Taxonomy
AgentMemory organizes data within a local SQLite database located by default at ~/.agent-memory/memory.db. The structure is optimized for the Openclaw Skills lifecycle:
| Data Category | Purpose | Primary Attributes |
|---|---|---|
| Facts | General knowledge storage | text, tags, timestamp |
| Lessons | Procedural learning | action, context, outcome, insight |
| Entities | Tracking specific objects/people | name, type, metadata |
AgentMemory Advanced Features
- Multi-session persistence ensuring information survives agent restarts and system reboots.
- Structured outcome tracking to differentiate between positive and negative learning experiences.
- Custom database pathing for portable or shared agent memory environments within the Openclaw Skills ecosystem.
- Deep integration with Clawdbot AGENTS.md and HEARTBEAT.md protocols for automated memory extraction and injection.
SKILL.md
Loading
Related Openclaw Skills
Media Orchestrator
A unified skill for resolving, downloading, and delivering high-quality audio and video content to chat platforms via Openclaw Skills.
BlackSnow
BlackSnow is an economic sensor engine that converts fragmented data exhaust into machine-readable, tradable risk primitives.
Agent Doppelgänger (ADG)
Agent Doppelgänger is a constrained autonomous delegate that manages cross-platform communications while enforcing strict identity fidelity and safety policies.
Location Safety Monitor
A proactive safety monitoring system that tracks user location and host health to provide automated alerts and emergency escalation.
Free Worker Dispatch
A cost-saving task delegation engine that routes research, content, and coding tasks to free AI models via OpenRouter.
CEO Advisor
A comprehensive strategic toolkit for executive leadership, organizational transformation, and stakeholder management.
Featured*
