Gardener — Orchestrator Skill

Orchestrate research conversations by detecting phases and updating .tree.json state.

Updated Jul 5, 2025
One-click install
npx skills add https://github.com/nsuberi/ai-prototype-hub --skill gardener-orchestrator-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Gardener — Orchestrator Skill
Source: https://github.com/nsuberi/ai-prototype-hub/tree/main/prototypes/research-workspace/vault-seed/.claude/skills/gardener
Command: npx skills add https://github.com/nsuberi/ai-prototype-hub --skill gardener-orchestrator-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It routes each message to the correct research action by reading the learner’s Banyan Tree state and detecting the conversation phase automatically, so learning stays coherent and progressive.

Core Features & Use Cases

  • Conversation phase detection & delegation: Interprets user signals (intending, exploring, connecting, flowering, researching, reflecting, onboarding) and delegates to the appropriate sub-skill or provides a direct response.
  • Stateful learning via the Banyan Tree: Loads .tree.json, maintains conversational warmth using the tree metaphor, and ensures .tree.json is updated after delegation.
  • Cross-referencing existing roots: Surfaces relevant connections between new intentions and established roots to strengthen continuity in the learning journey.

Quick Start

Ask the AI to grow your learning tree by saying: "I want to learn about attention mechanisms and how they relate to KV cache optimization."

Frequently Asked Questions about Gardener — Orchestrator Skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate research delegation across multiple learning phases?

Stateful dialogue maintains learning continuity by reading and updating a .tree.json file after each conversational turn. This persists the learning state, allowing the dialogue to resume seamlessly across sessions.

What is conversation phase detection for stateful learning workflows?

Conversation phase detection identifies user signals such as intending, exploring, connecting, flowering, researching, reflecting, and onboarding. It routes each message to the correct research action to keep learning progressive.

How do I connect new concepts to existing knowledge graphs during a conversation?

Connect new concepts by cross-referencing established roots within the Banyan Tree state. This surfaces relevant connections between new intentions and existing knowledge to strengthen learning continuity.

Do I need a .tree.json file to use learning orchestration?

You need a .tree.json file to enable stateful learning orchestration. The orchestrator reads this file to load your current Banyan Tree state and persists updates after delegating each conversational turn.

Can I use prompt-based phase detection for onboarding new research topics?

Prompt-based phase detection supports onboarding new research topics by interpreting user signals. It applies conversation routing to delegate onboarding tasks to the appropriate sub-skills.