Skills Plugins MCP Prompt Model 博客 我的中心

handoff

Clipboard-ready handoff prompt for another agent to investigate or continue a task.

DeepseekModel 官方收录技能 质量 优秀 · 90 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=openclaw-agent-skills-skills-handoff-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name handoff description Clipboard-ready handoff prompt for another agent to investigate or continue a task. Handoff Write a clipboard-ready prompt for another agent to investigate, discuss, or work on a specific task. Use when the user asks for handoff <task> , "write a handoff", "delegate this", or wants a prompt for another agent. Workflow Identify the task from the user text. If the user gives only a short label, infer from the current repo, recent discussion, branch name, linked issue/PR, docs, and obvious nearby context. Gather enough context to write a useful handoff: repo/product identity, relevant issue/PR/branch names, likely modules, constraints, and known symptoms. Do not perform the receiving agent's full independent review or decide the final technical direction for them. Write a standalone prompt for a fresh agent. Copy the full prompt to the clipboard. Final reply: terse confirmation with the task title. Do not paste the full prompt unless the user asks. Handoff Prompt Rules The prompt must: Start a discussion, not a command-only work order. Ask the receiving agent to do an extensive independent review before changing anything. Make clear that the receiving agent owns that review; the handoff only gives starting context and known constraints. Ask the agent to decide whether the task is a good idea, stale, already solved, over-scoped, or better handled differently. Assume the agent starts in the repo, a parent directory, a workspace directory, or a home directory and can find the repo itself. Avoid filesystem paths. No absolute paths, home-directory paths, checkout names, or repo-relative file paths unless the user explicitly requests them. Use portable anchors instead: repo owner/name, product/module names, issue/PR URLs, branch names, package/plugin names, public symbols, command names, config keys, exact error text, docs titles, and search terms. Include enough context for the receiving agent to get the right repo, boundary, and desired outcome. Include constraints, non-goals, validation expectations, and the desired output shape. Tell the receiving agent to re-check live repo/GitHub/CI state where relevant. Tell the receiving agent not to push, merge, close issues/PRs, label, or post public comments unless the handoff explicitly asks for it. Prompt Template Use this shape by default: I want to discuss and possibly work on: <short task title> Context: - <portable repo/product context> - <what triggered this task> - <known current state, branch/issue/PR names or URLs if relevant> - <important constraints and ownership boundaries> Before doing any implementation: - Find the right repository from the current directory, a parent directory, or the usual workspace. - Read the local agent/repo instructions. - Inspect the relevant code, docs, tests, recent commits, and linked issue/PR state. - Decide whether this task is still real, whether the proposed direction is a good idea, and whether a smaller/better fix exists. - Call out stale assumptions, hidden risks, and anything that should stop the work. Task: - <what to investigate or implement if the review supports it> - <expected behavior or decision criteria> - <non-goals> Validation: - <focused tests/checks/live proof expected> - <what evidence should be included> - <what is explicitly not required> Output: - Start with your review findings and recommendation. - Then give the proposed plan or patch summary. - If you edit code, keep changes scoped and report exact proof run. - Do not push, merge, close issues/PRs, label, or post public comments unless explicitly told. Clipboard On macOS: pbcopy < /tmp/handoff-prompt.txt Use a temp file or pipe. Avoid inline shell quoting for prompts containing backticks, $ , quotes, or user text. If pbcopy is unavailable, use the obvious platform clipboard tool ( wl-copy , xclip , clip.exe ) or print the prompt and say clipboard copy was unavailable. Quality Bar No invented facts. Mark reviewed facts as such only after checking them. No path leakage. Rewrite any accidental path as a symbol, module, command, issue/PR URL, or search term. Enough context for a fresh agent to orient; no giant brain dump. First real instruction to the receiving agent: review, discuss, assess.
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

验证码 --

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。