A conceptual framework providing AI agents with persistent, structured memory across episodic, semantic, procedural, and relational types.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-memory-succesion
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-succesion 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 Umnix AgentMemory Skill?
The Umnix AgentMemory skill is a comprehensive conceptual framework designed to transform AI agents from stateless tools into persistent, learning entities. By implementing a sophisticated memory architecture, agents can maintain context across sessions, accumulate institutional knowledge, and personalize interactions. Within the ecosystem of Openclaw Skills, this framework defines how agents store and retrieve information through four distinct memory types: episodic for events, semantic for facts, procedural for workflows, and relational for social graphs.
This skill serves as the blueprint for designing agents that understand their history and can transfer their expertise through succession protocols. It emphasizes vector-based semantic search, allowing agents to find relevant memories based on meaning rather than simple keyword matching. By adopting these principles, developers can build more reliable and intelligent agentic systems that grow more effective over time.
Umnix AgentMemory Skill Use Cases
- Designing long-term memory strategies for autonomous agents in enterprise environments.
- Implementing audit trails and event-based episodic records for regulatory compliance.
- Building domain-specific semantic knowledge bases for specialized technical support agents.
- Documenting and optimizing procedural workflows for complex multi-step automations.
- Managing user preferences and interaction patterns through relational memory for personalized experiences.
How Umnix AgentMemory Skill Works
- The agent captures significant events and interactions as timestamped episodic memories during operations.
- Raw text and concepts are converted into numerical vector embeddings to enable semantic meaning-based search.
- The system organizes factual data and domain expertise into a persistent semantic memory layer.
- Successful workflows and problem-solving strategies are formalized into procedural memory for future reuse.
- Agents retrieve contextually relevant information by querying the memory store based on current task metadata or user identity.
- Upon agent retirement, the succession protocol transfers the curated memory set to a designated successor agent.
Umnix AgentMemory Skill Setup
As a conceptual framework skill for Openclaw Skills, the implementation is handled via architectural design and the Umnix platform interface.
# Access the conceptual framework and design guidelines
# No direct CLI installation is required for this logic-only skill
# Visit umnix.in to configure agent memory settings
Umnix AgentMemory Skill Data Schema & Taxonomy
The AgentMemory skill organizes data into four primary taxonomies to ensure high retrieval relevance:
| Memory Category | Data Structure | Primary Purpose |
|---|---|---|
| Episodic | Timestamped logs and decision records | Audit trails and event history |
| Semantic | Factual knowledge and concepts | Domain expertise and user preferences |
| Procedural | Step-by-step workflow definitions | Process automation and best practices |
| Relational | Interaction patterns and social graphs | Personalization and relationship management |
Umnix AgentMemory Skill Advanced Features
- Vector-Based Semantic Search: Enables agents to find memories by conceptual meaning rather than exact word matches.
- Automated Succession Protocol: Facilitates the seamless transfer of institutional knowledge from retired agents to successors.
- Memory Lifecycle Management: Includes automated archival processes to deprioritize outdated information while maintaining auditability.
- Context-Aware Retrieval: Dynamically filters memory results based on the current task, user identity, and problem domain.
- Memory Isolation: Ensures data privacy by scoping memories strictly to specific agents or organizational boundaries.
SKILL.md
Loading
Related Openclaw Skills
AI Hallucination Correction (CN)
A specialized intent alias for VeriClaw designed to detect, diagnose, and fix AI hallucinations and false task completions through evidence-based correction loops.
Umnix Jury System
A conceptual framework skill for transparent, multi-perspective AI dispute resolution featuring a 12-juror parallel AI panel and human oversight.
Umnix Constitution Engine Skill
A foundational governance framework defining mandates, permissions, and performance standards for autonomous AI agents.
HPM Solver
A mathematical optimization tool designed to calculate the most efficient purchase combinations in Harry Potter Magic Awakened.
Google Sheets Maton Gateway
A powerful interface for AI agents to interact with Google Sheets via the Maton Gateway with full OAuth support.
Fal.Ai Media Generation
A powerful integration for generating professional-grade AI media including high-resolution images, cinematic videos, and accurate audio transcripts.
Featured*
