Skip to content
Back to skills

Claude Handoff

ASecurity

把当前对话交接给一个全新的后台 agent,让它立刻接手工作。

  • 408 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 29, 2026
ai-agentsapi

Works with

  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 29, 2026

npx -y skills add devcxl/mattpocock-skills-zh --skill claude-handoff --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Claude Handoff?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Claude Handoff
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/devcxl-claude-handoff/badge)](https://www.skillsdirectory.com/skills/devcxl-claude-handoff)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: claude-handoff
description: 把当前对话交接给一个全新的后台 agent,让它立刻接手工作。
argument-hint: "下一个会话将用来做什么?"
disable-model-invocation: true
---

为当前对话撰写一份交接摘要,让一个全新的 agent 能够继续工作。不要把摘要保存到本地,而是启动一个后台 agent,把这份摘要作为它的提示词注入:`claude --bg --name "<描述性名称>" "<交接摘要>"`。它会在当前工作目录启动并立刻返回;用户通过 `claude agents` 来管理它。

务必传 `-n`/`--name` 并配以描述性名称(例如 `--name "修复登录 bug"`):它会被设置为任务列表、会话选择器以及终端标题中所显示的标签。

在摘要里加一个 "suggested skills" 部分,建议下一个 agent 调用哪些技能。

不要重复已经被记录在其他产物(规格、计划、ADR、issue、commit、diff)中的内容。引用时用路径或 URL。

抹去任何敏感信息,例如 API 密钥、密码或可识别个人身份的信息:摘要会成为 agent 的提示词。

如果用户传入了参数,把它们当作对下一个会话关注点的描述,并据此裁剪摘要。

Files in this skill

  • SKILL.md1.2 KB
  • agents/openai.yaml141 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…