An interactive knowledge-mapping skill that helps users and AI agents co-create structured domain graphs to achieve targeted, systematic learning without information overload.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install learn-graph
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 learn-graph 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 Learn Graph?
Learn Graph is a specialized methodology designed to guide learners and AI agents through the process of systematically dissecting and mastering any complex new domain. Built on the core philosophy that the act of constructing a knowledge graph step-by-step is itself the most effective learning experience, this system rejects passive consumption in favor of active, guided co-creation. By avoiding pre-made, overwhelming templates, it keeps learners actively engaged, focused, and intellectually invested in building their own customized educational roadmap.
As a core component within the Openclaw Skills registry, this tool integrates seamlessly into workflows where developers and researchers need to quickly close knowledge gaps. Instead of dumping a massive web of complex definitions, it uses Socratic questioning to establish a structural skeleton of concepts, usages, and parent-child dependencies. This approach makes it easier to identify high-return areas of study and clear, common-sense entry points, changing the way agents and humans acquire domain-specific expertise.
Learn Graph Use Cases
- Systematically entering and mastering a completely new technical domain without getting lost.
- Overcoming learning anxiety when unsure where to start or how deep to dive into a subject.
- Resolving non-systematic learning gaps by visualizing relationships between foundational and advanced concepts.
- Mapping an efficient, customized learning path to achieve a specific, immediate practical goal.
- Quantifying domain scale to answer the critical learning question: 'How much study is enough?'
How Learn Graph Works
- Define Target and Purpose: Identify the specific domain X and the exact goal of learning (integrating with other Openclaw Skills like learn-occam to define the scope based on the actual problem to be solved).
- Co-Construct the Graph Skeleton: Build out the graph's skeleton focusing strictly on concepts, practical usage, and parent-child contextual relationships. The agent guides the user with step-by-step questions rather than dumping a completed graph.
- Identify High-Value and Entry Nodes: Mark 'reuse value' nodes (concepts with many dependencies that offer the highest learning ROI) and 'entry point' nodes (concepts that can be easily understood from common-sense knowledge).
- Generate the Actionable Path: Trace a clear learning route starting from the entry point, respecting all parent-child hierarchies. Tailor the granularity dynamically and stop once the path covers the nodes required to solve the initial purpose.
- Handover to Practice: Seamlessly transition from theoretical mapping to execution, bridging to skills like learn-prototype to build low-fidelity, practical prototypes based on the entry points identified.
Learn Graph Setup
To install and activate this skill within your workspace, execute the following command in your terminal. This will add the configuration to your Openclaw Skills library:
openclaw install learn-graph
Once installed, initialize the skill in your agent profile by referencing it in your active environment:
openclaw skill enable learn-graph
After activation, your AI agent will be equipped to systematically guide you through knowledge mapping sessions.
Learn Graph Data Schema & Taxonomy
The skill organizes learning domains using a hierarchical, contextual metadata schema. Below is the structured classification used to build out the learning path:
| Attribute | Description | Purpose |
|---|---|---|
| Concept / Name | The specific term, tool, or theory within domain X. | Core identity of the node |
| Usage | What the concept is used for and what real-world problem it solves. | Pragmatic grounding |
| Parent Node | What target objective or higher-level goal this node serves. | Downstream context |
| Child Node | What lower-level concepts, tools, or libraries this node relies on. | Upstream prerequisites |
| Reuse Value | A metric based on the number of parent nodes depending on this node. | Prioritizing high-ROI nodes |
| Entry Point | A binary flag indicating if a node is accessible via everyday common sense. | Defining the starting line |
Learn Graph Advanced Features
- Interactive Scaffolding: Prompts users with targeted questions to co-create graph skeletons rather than overwhelming them with massive pre-built diagrams.
- Knowledge Verification Safety: Operates under a strict 'confirmed knowledge only' rule, prompting the user directly with confirmation questions before assuming prerequisite familiarity.
- Dynamic Granularity Shifting: Enables smooth zooming in and out of the graph, from macro domain overviews down to fine-grained technical specifications.
- Cross-Skill Synergy: Integrates natively with complementary Openclaw Skills, allowing smooth handoffs to learn-occam for scope limitation and learn-prototype for immediate hands-on practice.
SKILL.md
Loading
Related Openclaw Skills
Self-Improvement Daemon
An autonomous background daemon that drives continuous system evolution, self-improvement, and recursive capability updates.
Self-Improvement Daemon
An AI agent skill designed to guide users through mounting virtual disk images (ISO) and installing classic retro games using Daemon Tools Lite.
Solve Problem
An intelligent AI agent skill designed to search knowledge bases and dynamically research comprehensive solutions to engineering and technical challenges.
Side Hustle
An automated search capability designed to help users identify side income, freelancing, and part-time gig opportunities via automated browser searches.
learn-prototype
An active learning agent skill that forces users to bypass analysis paralysis, build a raw prototype immediately, and iteratively refine it through guided self-correction.
learn-feynman
A structured Feynman technique workflow designed to test your understanding of complex concepts by prompting you to teach them simply to an inquisitive AI student.
Featured*
