claude-handoff

Launches a background Claude agent seeded with a handoff summary of the current conversation.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Fatih0234/mattpocock-skills-pi --skill claude-handoff-fatih0234
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-handoff
Source: https://github.com/Fatih0234/mattpocock-skills-pi/tree/main/skills/in-progress/claude-handoff
Command: npx skills add https://github.com/Fatih0234/mattpocock-skills-pi --skill claude-handoff-fatih0234

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Ending a long coding session loses context, and manually writing continuation notes is tedious. This Skill writes a handoff summary of the current conversation and immediately launches a fresh background agent seeded with that summary, so work continues without losing momentum. ## Core Features & Use Cases - Automatic Background Agent Launch: Runs claude --bg --name "<descriptive name>" with the handoff summary as the prompt, returning immediately while the agent works. - Context-Aware Summaries: References existing artifacts (specs, plans, ADRs, issues, commits, diffs) by path or URL instead of duplicating them, and suggests which skills the next agent should invoke. - Safe Handoffs: Redacts sensitive information such as API keys, passwords, and personally identifiable information before the summary becomes the agent's prompt. - Use Case: You finish a debugging session at the end of the day and want the fix implemented overnight. Invoke the skill with a note like "continue fixing the login bug" and a named background agent picks up the work in the current directory. ## Quick Start Ask the assistant to hand off the current conversation to a background agent focused on your next task, for example: hand off this session to a new agent that will finish the login bug fix.

Frequently Asked Questions about claude-handoff

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

FAQPage Schema
How do I hand off a Claude Code session to another agent?▼

Invoke the claude-handoff skill, optionally describing what the next session should focus on. It writes a handoff summary and launches a background agent with `claude --bg --name "<name>"` seeded with that summary as its prompt.

How do I run a Claude agent in the background?▼

Use `claude --bg --name "<descriptive name>" "<prompt>"`, which starts the agent in the current working directory and returns immediately. You manage running background agents with the `claude agents` command.

What should a handoff summary between AI agents include?▼

Include the current state of the work, key decisions, and a suggested skills section naming which skills the next agent should call. Reference existing artifacts like specs, plans, ADRs, issues, and diffs by path or URL instead of duplicating their content.

Does the handoff summary include secrets or API keys?▼

No. The skill explicitly redacts sensitive information such as API keys, passwords, and personally identifiable information, because the summary becomes the next agent's prompt and could otherwise expose credentials.

Can I control what the next background agent works on?▼

Yes. Any arguments you pass to the skill are treated as a description of what the next session will focus on, and the handoff summary is tailored accordingly.