AgentSkillsCN

handover

为我们的对话生成一份全面的 Markdown 摘要,以便无缝交接给另一位 AI 代理。适用于结束对话或移交工作任务时使用。

SKILL.md
--- frontmatter
name: handover
description: Generate a comprehensive markdown summary of our conversation for seamless handoff to another AI agent. Use when ending a session or transferring work.

Conversation Handoff Summary

Required Output

This command generates and saves a markdown file. Always:

  1. Analyze the conversation
  2. Create the file using Write tool
  3. Save to: ./handoff-summary-YYYYMMDD-HHMMSS.md

The purpose is file creation, not just information display.

Generate comprehensive summary for AI agent handoff: $ARGUMENTS

What This Command Does

This command creates a detailed markdown summary of our entire conversation, structured to enable another AI agent to seamlessly continue the work. The summary includes:

  • Context and Background: Initial problem statement and user requirements
  • Work Completed: Detailed list of all tasks accomplished
  • Technical Decisions: Key architectural and implementation choices made
  • Code Changes: Summary of files modified, created, or refactored
  • Current State: Where the project stands now
  • Pending Items: Any unfinished tasks or future considerations
  • Important Notes: Critical information for continuation

Output Format

The command generates a markdown file with the following structure:

markdown
# Project Handoff Summary

Generated: [timestamp]

---

## Overview

Brief description of the project and main objectives

## User Context

- User's technical background and preferences
- Specific requirements and constraints
- Communication style preferences

## Original Requirements

- Initial problem statement
- Key goals and objectives
- Success criteria

## Completed Work

### Task 1: [Description]

- What was done
- Why this approach was chosen
- Key code changes
- Files affected

### Task 2: [Description]

...

## Project Structure

Current state of the codebase:

- Directory structure
- Key files and their purposes
- Dependencies added/modified

## Technical Decisions

### Decision 1: [Topic]

- Options considered
- Chosen approach
- Rationale

### Decision 2: [Topic]

...

## Code Examples

Key code snippets demonstrating important implementations

## Current State

- What's working
- What's being worked on
- Known issues or limitations

## Pending Tasks

- [ ] Task 1
- [ ] Task 2
- [ ] Task 3

## Important Warnings

- Critical information to avoid breaking changes
- Security considerations
- Performance implications

## Next Steps

Recommended actions for continuing the work

## Related Files

- File 1: Purpose and recent changes
- File 2: Purpose and recent changes

## Additional Context

Any other relevant information for smooth continuation

Benefits of This Summary

  • Continuity: New agent can pick up exactly where we left off
  • Context Preservation: All important decisions and rationale are documented
  • Efficiency: Reduces need to re-explain or rediscover information
  • Clarity: Structured format makes information easy to find
  • Completeness: Captures both technical and conversational context

Usage Examples

Basic usage:

code
/handover

With specific focus:

code
/handover focusing on authentication implementation

For specific date range:

code
/handover for work done today

Key Sections Explained

User Context

Captures communication preferences, technical level, and any specific requirements mentioned during conversation

Technical Decisions

Documents why certain approaches were chosen over others, preventing future agents from undoing intentional choices

Code Examples

Includes actual code snippets for complex implementations, ensuring the next agent understands the implementation style

Important Warnings

Highlights any critical information that could cause issues if not known (e.g., "Don't modify X because it will break Y")

Related Files

Lists all files that were created or modified, with brief descriptions of their purpose and changes

Notes for the Next Agent

The summary will include:

  • Conversation tone and style preferences
  • Any tools or commands created during the session
  • User's stated preferences (e.g., no emojis, no conventional commit prefixes)
  • Current working directory and environment setup
  • Any external dependencies or API keys mentioned

Output Options

The command will:

  • Generate a comprehensive markdown file
  • Save it as handoff-summary-[timestamp].md
  • Display the content for review
  • Optionally include conversation transcript excerpts for critical decisions

This ensures perfect continuity when switching between AI agents or resuming work after a break.