handoff

Save and resume task context across conversations via a CLI script.

45|50|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/zocomputer/skills --skill handoff-zocomputer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: handoff
Source: https://github.com/zocomputer/skills/tree/main/Community/handoff
Command: npx skills add https://github.com/zocomputer/skills --skill handoff-zocomputer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill allows an AI to pause a task, request user input, and resume the task in a future conversation, ensuring continuity even if the user responds through a different channel or at a later time.

Core Features & Use Cases

  • Pause and Resume: Enables the AI to save its current state and a specific question for the user.
  • Cross-Conversation Continuity: Ensures that user responses are captured and acted upon in subsequent interactions.
  • Use Case: An AI is processing a complex report but needs clarification on a specific data point. It saves the report's status and asks the user for clarification, then resumes processing once the user provides the answer in a new chat session.

Quick Start

Run python3 /home/workspace/Skills/handoff/scripts/handoff.py check to see if there is a pending handoff.

Frequently Asked Questions about handoff

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

FAQPage Schema
How do I pause an AI agent task and resume it in a different conversation?

To pause an AI agent task and resume it later, you use a cross-conversation continuity system that saves the current state and user questions to a persistent JSON file. The handoff mechanism stores the task context so a new session can retrieve and continue processing exactly where the previous one left off.

What is cross-conversation continuity in AI agent state management?

Cross-conversation continuity in AI agent state management is the ability to maintain task context across separate chat sessions. It works by saving the agent's current state and pending user questions to a JSON file, allowing subsequent interactions to load the context and resume the workflow seamlessly.

How do I check for pending user input questions when starting a new AI conversation?

To check for pending user input questions when starting a new AI conversation, run the handoff check command via CLI. This script inspects the JSON state file for any saved task context or questions left by a previous session, allowing the agent to resume the conversation flow immediately.

Can I save AI task context and clear handoff states using a CLI script?

Yes, you can save AI task context and clear handoff states using a CLI script. The system provides command-line operations to write the current state and user questions to a JSON file for persistence, or to clear the saved handoff state once the task is successfully resumed and completed.

Do I need a JSON state file for AI conversation flow persistence?

Yes, a JSON state file is required for AI conversation flow persistence. The handoff system relies on this file to store and retrieve the saved task context and user questions, ensuring that the pause-and-resume mechanism functions correctly across different conversation channels and sessions.

When should I not use a pause-and-resume system for AI agent continuity?

You should not use a pause-and-resume system for AI agent continuity if your conversations are strictly synchronous and never span multiple sessions or channels. It is designed for complex workflows requiring user input later, so simple single-session tasks without pending questions do not need handoff state management.