handoff

Detect user role and state to execute the next handoff action.

Updated Mar 23, 2026
One-click install
npx skills add https://github.com/jblacketter/tokenbench --skill handoff-jblacketter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: handoff
Source: https://github.com/jblacketter/tokenbench/tree/main/.claude/skills/handoff
Command: npx skills add https://github.com/jblacketter/tokenbench --skill handoff-jblacketter

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automatically coordinates AI handoff between roles by detecting the active role and current state, then executing the appropriate next action to keep workflows moving.

Core Features & Use Cases

  • Role-aware handoff orchestration: auto-detects lead or reviewer and advances the cycle.
  • State-driven decision making: reads the current handoff-state.json and cycle file to determine next steps.
  • Safe recovery and guidance: includes error-recovery rules and next-step prompts to preserve progress.

Quick Start

Run a handoff cycle by verifying your role and state, then execute the next step in the active cycle.

Frequently Asked Questions about handoff

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

FAQPage Schema
How do I automate AI workflow handoff between team roles?

Automate AI workflow handoff by detecting the active user role and current state, then executing the appropriate next action to keep collaborative cycles moving. The system reads handoff-state.json and active cycle files to determine the correct step automatically.

What is state-driven cycle management for collaborative AI workflows?

State-driven cycle management reads ai-handoff.yaml, handoff-state.json, and the active cycle file to determine the next step in an AI workflow. It maintains progress by preserving the current state across role transitions and plan or implementation cycles.

How do I recover from errors during an AI handoff cycle?

Recover from errors during an AI handoff cycle using built-in error-recovery rules and next-step prompts. The system reads the current handoff-state.json to safely guide the workflow back to the correct action without losing cycle progress.

Does AI handoff auto-detection work for both planning and implementation cycles?

AI handoff auto-detection works across both plan and implementation cycles. It identifies whether the active role is a lead or reviewer and advances the cycle accordingly by reading the active cycle file to execute the correct next step.

Do I need specific configuration files to run an automated AI handoff?

You need ai-handoff.yaml, handoff-state.json, and an active cycle file to run an automated AI handoff. The system reads these files to detect your role and state before executing the next action in the collaborative workflow.