A hybrid JSON and Markdown system for managing AI agent knowledge, user profiles, and automated task rhythms.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install lightweight-kb
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 lightweight-kb 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 Lightweight Knowledge Base & Task Manager?
The Lightweight Knowledge Base skill is a sophisticated memory management framework designed for AI agents. By utilizing a hybrid storage model that combines the queryable power of JSON with the human-readability of Markdown, it allows agents to maintain a three-layer architectural approach to information. This architecture moves from core specifications (the constitution) to execution methodologies (regulations) and finally to specific case nodes (knowledge).
This skill is a vital component of Openclaw Skills as it enables agents to evolve daily. It features an automated optimization process that consolidates new memories, updates user trait weights, and refines task execution logic without human intervention, ensuring the agent remains aligned with user preferences and goals.
Lightweight Knowledge Base & Task Manager Use Cases
- Automating recurring daily and weekly tasks through a standardized rhythm engine.
- Creating and maintaining detailed user profiles to personalize agent responses based on efficiency traits and communication preferences.
- Managing complex organizational knowledge by indexing Markdown files with structured JSON metadata.
- Implementing a self-evolving memory system that extracts methodologies from historical task execution cases.
How Lightweight Knowledge Base & Task Manager Works
- The system initializes a structured directory containing JSON templates for profiles, task rhythms, and knowledge indices.
- The agent monitors user interactions to dynamically update traits such as efficiency orientation and architectural sensitivity within the user profile.
- Scheduled routines are triggered based on the task rhythm table, handling everything from morning briefings to deep-dive weekly reviews.
- During the daily evolution phase, typically at midnight, scripts automatically merge new memory fragments and consolidate case studies into the permanent knowledge base.
- The agent utilizes the communication guidelines to ensure all outputs are data-driven, direct, and warm, effectively hiding internal processing while focusing on actionable conclusions.
Lightweight Knowledge Base & Task Manager Setup
To get started with this skill within your Openclaw Skills environment, initialize the knowledge base structure using the following commands:
# Initialize the directory structure and default templates
bash skills/lightweight-kb/scripts/init.sh
# Query the user profile to verify setup
bash skills/lightweight-kb/scripts/query.sh profile "efficiency"
# Manually trigger a daily evolution cycle
bash skills/lightweight-kb/scripts/daily_evolve.sh
Lightweight Knowledge Base & Task Manager Data Schema & Taxonomy
The skill organizes data across several specialized directories and files to ensure high performance and clarity:
| Path | Purpose |
|---|---|
data/user_profile.json |
Stores structured user traits, timezones, and preference weights. |
data/task_rhythm.json |
Contains the schedule for daily and weekly automated actions. |
data/kb_index.json |
Serves as the master index for all knowledge nodes. |
memory/kb/nodes/ |
Holds specific JSON nodes for personas, task executions, and methodologies. |
references/ |
Stores Markdown-based communication and task execution guidelines. |
Lightweight Knowledge Base & Task Manager Advanced Features
- Automated Midnight Evolution: A self-optimizing routine that performs knowledge graph updates and methodology extraction.
- Multi-Layered Task Classification: Differentiates between routine tasks, spontaneous daily tasks, and real-time communication responses.
- Dynamic Trait Weighting: Automatically adjusts user preference scores based on long-term interaction patterns.
- Direct-to-Conclusion Communication: Implements a specific output style that focuses on data-driven results while suppressing internal reasoning logs.
- Cross-Skill Integration: Designed to work seamlessly with other Openclaw Skills like memory managers and communication tools.
SKILL.md
Loading
Related Openclaw Skills
Facebook Ads Manager
A professional AI agent skill for orchestrating Meta Ads planning, operation, and scaling with strict safety protocols and mathematical precision.
Recruitly CRM MCP Skill
A powerful MCP-based integration that enables recruiters to manage candidates, job pipelines, and company data through natural language.
Add Feishu Channel
A seamless integration for Feishu and Lark that enables AI agent communication via a secure WebSocket connection.
OpenFunderse Strategy Skill
An advanced AI agent skill for proposing and submitting gated trade intents based on market snapshots and risk policies.
One-Click Task Dashboard
An automated visualization suite that generates a persistent, local web dashboard to monitor OpenClaw and macOS system tasks.
Fuego: Solana Agent Wallet
A local-first Solana wallet and infrastructure layer designed for AI agents to perform secure transfers, swaps, and automated purchases.
Featured*
