Skill Ontology
Skill Ontology organizes individual skills into a structured, composable network, enabling agents to reason, plan, and execute complex tasks as an extensible, maintainable capability system.
How it worksWhen querying for a task, the system traverses this graph to identify the necessary collections and skills to construct a capable agent.
Three-Layer Design
How skills are organized from abstract taxonomy to deployable packages.
1. Skill Taxonomy
The top layer defines the broad categorization and detailed tags of skills. It organizes capabilities into categories such as:
- Development, AI & Generation, Testing
- Research, Data & Science
- Business, Productivity, Security
2. Skill Relation Graph
The middle layer instantiates specific skills and defines how they interact. It maps relationships using edges like:
- compose_with: Combining patterns
- similar_to: Mapping alternatives
- depend_on: Establishing prerequisites
- belong_to: Sub-component within a larger skill
3. Skill Collection
Groups related skills into deployable units. These are the actual functional toolkits agents load at runtime. Examples:
Relationship Types
similar_toFunctionally equivalent or substitutable skills
belong_toHierarchical categorization within larger workflows
compose_withSkills that combine together with output-to-input flow
depend_onPrerequisites and environment setup requirements
Explore the skill network
Discover how AI skills connect, compose, and depend on each other across the platform.