A persistent, searchable memory layer that prevents AI agents from forgetting user preferences and project context between sessions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install persisent-mind
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 persisent-mind 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 PersistentMind?
PersistentMind provides a local-only, zero-dependency memory framework designed specifically for AI agents. By integrating this tool into your workflow, agents can store and recall vital information ranging from coding preferences to complex project architectures, ensuring that context is never lost across different chat sessions or development environments. This system allows developers to build a robust knowledge base using Openclaw Skills, effectively ending the cycle of re-explaining project requirements or repeating past corrections.
It operates entirely on your local machine, prioritizing privacy while offering powerful features like relevance-based search and importance scoring. Whether you are managing complex infrastructure or standardizing team coding styles, PersistentMind ensures your agent gets smarter over time by retaining every critical detail without the need for external API keys or cloud storage.
PersistentMind Use Cases
- Persisting user coding style preferences and architectural decisions across multiple AI sessions.
- Storing project-specific documentation and database schemas for instant agent recall during development.
- Tracking past mistakes and linter corrections to prevent the agent from repeating the same errors in future prompts.
- Creating a shared and portable team knowledge base for collaborative AI development and onboarding.
How PersistentMind Works
- Initialize the PersistentMind instance with a specific project scope and local storage path on your machine.
- Store facts, procedures, or corrections using the remember method, assigning importance scores and custom tags for better organization.
- Execute the recall function to perform a full-text search across stored memories, using relevance scoring to find the most pertinent information.
- Generate a formatted context block using get_context and inject it directly into AI prompts to provide the agent with immediate background knowledge.
- Maintain the memory library by updating existing entries, archiving stale data, or using the consolidation feature to merge duplicate memories.
PersistentMind Setup
To get started with this addition to your Openclaw Skills library, initialize the memory system within your Python project. Data is stored locally in the .persistentmind/ directory by default.
# Initialize PersistentMind in your environment
# Data storage is local-only with zero dependencies
from persistentmind import PersistentMind
# Create a project-specific memory instance
mm = PersistentMind(project="my-app-name")
PersistentMind Data Schema & Taxonomy
PersistentMind organizes data into a structured local hierarchy, ensuring fast retrieval and clear categorization. Memories are stored as local JSON objects categorized by type and scope.
| Field | Description |
|---|---|
| Memory Types | Supports fact, preference, procedure, correction, context, relationship, and reminder |
| Memory Scopes | global (universal rules), project (specific to a repo), and session (temporary context) |
| Metadata | Includes importance scores (1-10), auto-extracted tags, source identification, and expiration dates |
| Storage Path | Defaults to a local .persistentmind/ directory within the project root |
PersistentMind Advanced Features
- Automated memory consolidation using near-duplicate detection to keep your context window clean and efficient.
- Importance-based ranking logic that ensures critical corrections and security rules always surface first in prompts.
- Portability features including JSON export and import for sharing curated memory sets across distributed development teams.
- Intelligent auto-expiry settings that automatically prune temporary tokens or stale session data after a defined period.
- Advanced filtering capabilities by memory type, importance threshold, or custom tags to fine-tune context injection.
SKILL.md
Loading
Related Openclaw Skills
Adaptive Learning Agent
A zero-dependency Python tool for AI agents to capture, store, and retrieve real-time learnings and error resolutions.
Multilingual Greetings
A specialized skill for AI agents to deliver culturally appropriate, multilingual greetings across formal and informal contexts.
OpenReview Review Analyzer
An automated research tool that fetches, aggregates, and synthesizes peer reviews from OpenReview into structured analytical reports.
Weather Open-Meteo
A high-performance weather skill providing real-time atmospheric data and multi-day forecasts via Open-Meteo and wttr.in without requiring API keys.
Prompt Performance Tester
A model-agnostic benchmarking tool that compares performance, cost, and quality across nine major AI providers simultaneously.
TokenWatch
TokenWatch is a comprehensive spending monitor that helps developers track and optimize AI token costs across multiple providers in real-time.
Featured*
