handoff

Generate an end-of-session audit and cold-start handoff prompt.

1|Updated May 15, 2026
One-click install
npx skills add https://github.com/mslshao/claude-harness --skill handoff-mslshao
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: handoff
Source: https://github.com/mslshao/claude-harness/tree/main/dotclaude/skills/handoff
Command: npx skills add https://github.com/mslshao/claude-harness --skill handoff-mslshao

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents important session context, workflow updates, and in-flight work from being lost when a conversation ends. It gives you a structured way to audit your personal AI setup against what you learned in the session and then turn the current state into a clean cold-start prompt for the next session.

Core Features & Use Cases

  • Two-phase session wrap-up: Audits configs, hooks, agents, skills, and memory patterns before generating a ready-to-paste handoff prompt.
  • Cross-surface state gathering: Pulls in-flight context from beads, GitHub PRs, worktrees, and optionally Slack, Jira, and Confluence so the next session starts with current reality instead of partial memory.
  • Action gating for changes: Separates trivial fixes that can be applied immediately from substantive changes that should be proposed for user sign-off.
  • Use case: After a long coding or planning session, use this Skill to capture active work, calibration changes, open asks, and adjacent responsibilities so a new session can resume without re-deriving context.

Quick Start

Ask the AI to run the handoff skill for this session and generate both the audit and the copy-paste-ready cold-start prompt.

Frequently Asked Questions about handoff

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

FAQPage Schema
How do I preserve AI session context and generate a handoff prompt for my next session?

To preserve AI session context, you can generate an end-of-session handoff prompt that audits configuration changes and captures in-flight work from GitHub, worktrees, and Slack. This creates a copy-paste-ready cold-start prompt for your next session.

What is the best way to audit personal AI configuration before ending a coding session?

The best way to audit personal AI configuration is to perform a two-phase session wrap-up that reviews hooks, agents, skills, and memory patterns. This separates trivial fixes from substantive changes requiring sign-off before generating a handoff prompt.

How does cross-tool state capture work for GitHub PRs and Jira tickets during session wrap-up?

Cross-tool state capture works by pulling in-flight context from GitHub PRs, worktrees, Jira, Confluence, and Slack during session wrap-up. This gathers current reality across surfaces so the next session resumes with accurate, up-to-date context.

Can I use a session summary skill to sync workflow alignment across beads and Confluence?

Yes, you can use a session summary skill to sync workflow alignment across beads and Confluence. It captures active work and adjacent responsibilities, synthesizing them into a concise reusable prompt for the next session.

Does the handoff process support gated application of trivial versus substantive configuration changes?

Yes, the handoff process supports gated application of changes by separating trivial fixes that can be applied immediately from substantive calibration changes that are proposed for user sign-off before being committed.

When do I need to generate a cold-start prompt for cross-tool state capture?

You need to generate a cold-start prompt when ending a long coding or planning session where important context, workflow updates, and in-flight work across multiple tools would otherwise be lost without a structured handoff.