Skill Mastery and Prerequisites
The skills system in dLANDiscord includes a hierarchical dependency model, ensuring that agents only utilize complex capabilities once their prerequisites are met.
Evolution and Purpose
Implemented in Git commit 229707c3, the "Skill Mastery Branches" were added to bring more structure to autonomous swarms. It prevents agents from attempting high-level tasks (like "Cloud Architecting") without first validating the foundational environment (like "Linux Sysadmin").
Key Features
- Skill Prerequisites: Manifests can now define a list of
prerequisite_skillsthat must be active before the target skill can be injected. - Mastery Levels: Skills are categorized into levels (Foundation, Intermediate, Advanced, Mastery).
- Dynamic Validation: The
skills.gosystem recursively checks the dependency tree duringInjectSkill.
Manifest Example
{
"skill_id": "api-development",
"name": "API Development",
"category": "backend",
"prerequisite_skills": ["go_web_backend"],
"injections": {
"prompt_suffixes": ["Focus on RESTful patterns and documentation."]
}
}
Mermaid Diagram: Skill Dependency Tree
graph TD
Base[Linux Sysadmin] --> Go[Go Web Backend]
Go --> API[API Development]
Go --> DB[Database Design]
API --> Master[Master Backend Architect]
DB --> Master
Guidance for AI Agents
- Identify Blockers: If a skill injection fails due to missing prerequisites, inform the user which foundational skill needs to be added first.
- Progressive Learning: When starting a new project, prioritize foundational skills to build a robust dependency tree for later, more complex tasks.