orchestrator-state

Persist and coordinate AI workflow state across subtasks with hot and history files.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/KingSharkG/ai-agents-workflow --skill orchestrator-state
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: orchestrator-state
Source: https://github.com/KingSharkG/ai-agents-workflow/tree/main/skills/orchestrator-state
Command: npx skills add https://github.com/KingSharkG/ai-agents-workflow --skill orchestrator-state

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

The orchestrator-state skill outlines how to store and manage persistent runtime and history data for the AI workflow so subtasks can be dispatched, tracked, and reconciled reliably.

Core Features & Use Cases

  • Centralizes hot state (orchestration-state.json) and history (orchestration-history.json) to coordinate subtask transitions.
  • Enforces phase transitions, gating, and post-approval closure to ensure consistent task progress and auditability.
  • Supports migration from legacy v1 state to v2 with safe upgrades and history split.

Quick Start

Initialize the task state by creating orchestration-state.json and optionally orchestrations-history.json, then dispatch the first subtask.

Frequently Asked Questions about orchestrator-state

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

FAQPage Schema
How do I maintain persistent AI workflow state across multiple subtasks?

You can maintain persistent AI workflow state by centralizing hot execution data in orchestration-state.json and long-term history in orchestration-history.json to coordinate subtask transitions reliably.

What is the best way to implement phase gating and post-approval closure in AI pipelines?

Implement phase gating and post-approval closure by enforcing strict phase transitions within the orchestrator state, ensuring consistent task progress and full auditability across multi-subtask pipelines.

How do I migrate legacy v1 state files to v2 safely?

You can migrate legacy v1 state to v2 safely using the built-in migration mechanism, which performs safe upgrades and splits runtime state from long-term history files.

Does AI subtask dispatch require separate hot and history files for fault recovery?

Yes, separating hot execution state from long-term history files enables robust tracking, auditing, and fault recovery by applying atomic updates and P1/P4 governance to multi-subtask pipelines.

How do I initialize task state to dispatch the first subtask?

Initialize task state by creating the orchestration-state.json file and optionally orchestration-history.json, which establishes the baseline required to dispatch the first subtask in the workflow.