follow-up-orchestrator

Evaluate multi-step turns and return a structured completion or blocking verdict.

2|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/metric-space-ai/ctox --skill follow-up-orchestrator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: follow-up-orchestrator
Source: https://github.com/metric-space-ai/ctox/tree/main/skills/system/mission_orchestration/follow-up-orchestrator
Command: npx skills add https://github.com/metric-space-ai/ctox --skill follow-up-orchestrator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Evaluate end-of-turn decisions by ensuring non-silent abandonment of meaningful work, producing a clear verdict on completion, blocking, or requirement for replanning, while preserving durable knowledge in the CTOX runtime store.

Core Features & Use Cases

  • End-of-turn evaluation that yields a compact structured outcome (done, needs_followup, blocked_on_user, blocked_on_external, needs_replan).
  • Follow-up generation: when needed, proposes a durable self-work or plan artifact to continue work, ensuring traceability and accountability.
  • Durable context awareness: integrates with runtime store, tickets, plans, and verification data to keep knowledge current across turns.

Quick Start

Run the end-of-turn evaluation to decide the current task state and surface a concrete follow-up when needed.

Frequently Asked Questions about follow-up-orchestrator

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

FAQPage Schema
How do I decide if a multi-step workflow turn is complete or needs a follow-up?

To decide if a multi-step workflow turn is complete or requires a follow-up, evaluate the ongoing turn against runtime data to return a structured verdict like done, needs_followup, or needs_replan. This prevents silent abandonment of meaningful work.

What is end-of-turn evaluation in durable workflow orchestration?

End-of-turn evaluation in durable workflow orchestration is the process of assessing an active turn to produce a stable outcome contract, ensuring durable runtime state, tickets, and plans are updated to enable explicit handoffs and traceability across turns.

How do I generate structured follow-up tickets when a task is blocked on external dependencies?

Generate structured follow-up tickets when a task is blocked on external dependencies by returning a blocked_on_external verdict and proposing durable follow-up data for persistence. This ensures traceability and explicit handoffs for the blocked workflow state.

When should I use an explicit replan verdict instead of marking workflow planning as done?

Use an explicit needs_replan verdict instead of marking workflow planning as done when the current plan is no longer viable and requires replanning. This enforces a stable output contract and preserves durable knowledge in the runtime store.

Can I persist durable state and knowledge across multiple turns during ticketing orchestration?

You can persist durable state and knowledge across multiple turns during ticketing orchestration by integrating the evaluation with the runtime store, tickets, plans, and verification data. This keeps knowledge current and enables explicit handoffs.