A sophisticated cognitive architecture tool designed to transform fragmented information into a structured, actionable external brain.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install context-manager-v1
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 context-manager-v1 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 Context Manager?
Context Manager is a specialized skill for AI agents that focuses on the philosophy that knowledge management is actually personal context management. Based on the Karpathy LLM Wiki principles, it assists users in synthesizing personal knowledge, experiences, emotions, and judgments into a cohesive cognitive framework. This tool moves beyond simple storage by helping users bridge the gap between external information and internal decision-making.
By leveraging Openclaw Skills, this system creates a digital mirror of your thought process. It allows you to build mind trees that automatically correlate external data from platforms like Feishu or WeChat with your own internal insights. The goal is to ensure that your knowledge base isn't just a digital graveyard of bookmarks, but a living, growing cognitive map that supports deep thinking and better decision-making.
Context Manager Use Cases
- Organizing fragmented information from social media and professional communication apps into a central context library.
- Recording internal cognitive logs that capture the specific triggers and judgments behind personal insights.
- Building automated mind trees to visualize the evolution of your understanding over time.
- Executing decision backtracks to review historical context before making high-stakes choices.
- Defining clear boundaries for AI assistance to ensure that critical thinking remains a human-driven process.
How Context Manager Works
- Multi-Source Collection: The skill gathers data from diverse inputs including chat records, documents, and manual cognitive logs.
- Semantic Anchoring: Information is categorized using a system of meaning-based tags such as growth pain points or decision backgrounds.
- Context Bridging: The agent creates logical connections between external data points and internal thoughts to show how insights were formed.
- Mind Tree Generation: The system synthesizes these connections into a visual or structured map, highlighting cognitive updates and new realizations.
- Decision Recall: During new scenarios, the skill retrieves at least two related historical context points to provide a baseline for objective analysis.
- Cognitive Maintenance: Periodic reviews within Openclaw Skills help prune redundant information and refine the core cognitive architecture.
Context Manager Setup
To deploy this skill, ensure your AI agent environment is active and follow these steps:
# Navigate to your agent's skill directory
cd my-ai-agent/skills
# Install the Context Manager skill
openclaw install context-manager
# Configure your integration tokens for external platforms
# (e.g., Feishu, WeChat) in the config.yaml file
Ensure that the description field in your agent configuration includes the necessary trigger words like manage context or cognitive map.
Context Manager Data Schema & Taxonomy
The Context Manager organizes data using a structured metadata approach to ensure high retrieval accuracy within Openclaw Skills:
| Property | Description | Type |
|---|---|---|
type |
The category of entry (e.g., Cognitive Log, Decision Backtrack) | String |
trigger_point |
The specific event or thought that initiated the entry | String |
tags |
Semantic anchors like #RelationshipAnchor or #CognitiveConflict | Array |
judgement |
The core human insight or decision made | String |
ai_analysis |
AI-generated pattern recognition and emotional sentiment | Object |
links |
Internal wiki-style links to related context files | Array |
Context Manager Advanced Features
- Automated Cognitive Mapping: Generates comprehensive mind trees for specific themes over a 3-month rolling window.
- FOMO Detection: Recognizes patterns of emotional decision-making in historical logs and issues warnings during investment or career choices.
- Dynamic Boundary Setting: Allows users to specify which cognitive tasks are strictly human-only and which can be delegated to the agent.
- Quarterly Pruning: An automated workflow to identify and remove redundant information that no longer serves the current cognitive framework.
- Cross-Platform Bridging: Seamlessly connects a Feishu document to a WeChat conversation and a personal insight in a single logical thread.
SKILL.md
Loading
Related Openclaw Skills
Guochuangsai Business Model Design Expert
An AI-powered consultant designed to build, optimize, and present winning business models for the China International College Students' Innovation Competition.
Agent Android
A local-first automation skill for controlling Android devices over a network without requiring USB connections, ADB, or root access.
LLM Memory Integration
A hybrid memory management system providing vector search and SQLite-based FTS for persistent AI agent context.
Duru Memory
A high-traceability memory continuity system for AI agents that uses structured Markdown files as a persistent source of truth.
WeChat Official Account Draft Publisher
A specialized automation skill for uploading Markdown or HTML content directly to the WeChat Official Account draft box.
Discord Voice
A high-performance voice integration that allows users to have natural, real-time spoken conversations with Claude AI within Discord voice channels.
Featured*
