handoff

审查 · 64
已收录

Produce a decision-safe handoff for a fresh agent. Use when a task changes hands, a session is ending midstream, a user asks for a handoff, or an agent needs to preserve a design/map/implementation decision without making the next agent rediscover it.

Verified installs0
Stars18
版本1.0.0
质量59/100 · 有潜力
信任64/100 · 仅限沙盒
审计76/100 · 需审查

供给资产档案

编程与开发 Agent

代码审查、仓库分析、测试、CI、GitHub、DevOps 与开发工作流 Skill。

浏览赛道

场景

GitHub automation

I need my agent to triage GitHub issues, review pull requests, and summarize repository changes.

适配 Agent

Claude Code + CLI + Codex

适用于 Codex、Claude Code、Cursor、CLI 或自定义 Agent。

安装

就绪

npx skills add ConnorGriffin/skills --skill handoff

维护状态

新鲜

距上次推送 4 天

风险

需审查

Financial research output is not financial advice; require human review before any live investment decision

GitHub 质量

18

59/100 质量 · 72/100 信任

覆盖标签

编程GitHub automation设计与创意agent-skill

审查说明

Financial research output is not financial advice; require human review before any live investment decision · Repository license is NOASSERTION; unclear licensing may hinder adoption.

Agent 采用评分卡

一眼查看信任、审计与安装准备度

这些分数综合公开仓库元数据、OpenAgentSkill 审查信号、维护新鲜度与安装准备度。它用于候选筛选,不替代人工审查。

质量

有潜力
59

有用的候选项,但采用前应与替代方案比较。

信任

仅限沙盒
64

有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。

审计

需审查
76

对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。

OpenAgentSkill 信任评分 v5

安装前需人工审查

仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。

CodexClaude CodeCursorOpenAgentSkill CLI

Stars

18 个 GitHub Stars

仓库活跃度

18 个 Star,5 个 Fork

维护状态

距上次推送 4 天

许可证

NOASSERTION

安装

npx skills add ConnorGriffin/skills --skill handoff

安装安全性

标准软件包或运行时安装路径

权限范围

文件系统或文档访问

Agent 结果

暂未有 Agent 结果数据

文档

README/SKILL.md 上下文充分

风险摘要

生产前审查

  • Repository license is NOASSERTION; unclear licensing may hinder adoption.
  • Financial research output is not financial advice; require human review before any live investment decision.
  • Low GitHub adoption signal
  • Quality score needs review

安装准备度

安装路径可用

  • 安装路径可用
  • 仓库证据可用
  • 已声明许可证
  • 暂无 Agent 验证结果证据

Agent 可读元数据

这个 Skill 的机器可读决策数据。

使用此区块或内嵌 JSON 判断 Agent 是否应安装该 Skill、选择替代方案,或先请求人工审查。

打开 JSON

适用任务

  • GitHub automation 工作流
  • Claude Code 团队
  • builders willing to evaluate younger projects
  • Inspect repository metadata

适用 Agent

CodexClaude CodeCursorOpenAgentSkill CLICLI

安装决策

命令
npx skills add ConnorGriffin/skills --skill handoff
策略
审查
人工审查

信任与风险

信任
64/100
审计
76/100
风险级别
需审查

结果闭环

端点
/api/agent/outcome
事件 ID
resolve
结果
5

安装命令

npx skills add ConnorGriffin/skills --skill handoff

不适用场景

  • 需要厂商支持 SLA 的团队
  • production agents without a repository review
  • Low GitHub adoption signal
  • Repository license is NOASSERTION; unclear licensing may hinder adoption.
  • 暂未有 OpenAgentSkill 使用反馈数据

Agent 安全 v2

60/100 · 安装前审查

已审查并附权限说明审查

可用候选,但 Agent 在安装前应展示权限与审计说明。

在真实工作区安装前需要人工批准。

通过 API 解析

网络访问

Skill 可能访问远程页面、API、仓库或外部服务。

文件系统访问

Skill 可能读取或写入项目文件、文档、生成产物或本地工作区状态。

  • Financial research output is not financial advice; require human review before any live investment decision

安装目标

在你的 Agent 工作流中安装此 Skill

通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。

skill install

OpenAgentSkill CLI

Resolve policy, run the source installer safely, and report a verified install receipt.

$ npx --yes https://github.com/Leon-Drq/openagentskill/releases/download/cli-v0.2.1/openagentskill-0.2.1.tgz install connorgriffin-handoff

Agent 解析计划

让 Agent 在安装前验证匹配度。

Resolve API 返回首选 Skill、替代方案、安全策略、审计说明、安装目标和可直接执行的提示词,无需抓取此页面。

打开文本计划

Agent 应检查

  • 从 Resolve API 检查任务匹配与替代方案。
  • 检查审计评分、信任评分和安全策略警告。
  • 检查 Codex、Claude Code、Cursor 或 CLI 的安装目标兼容性。

复制提示词

Task: Use handoff in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20handoff%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/connorgriffin-handoff/install
Install command: npx skills add ConnorGriffin/skills --skill handoff
Before running it, summarize audit warnings, required permissions, and the fallback skill if install is risky.

Agent 交接

把安装路径交给 Agent,而不是再给一个目录页。

通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。

打开安装 API

Agent 提示词

Use handoff for this task. Review https://www.openagentskill.com/api/skills/connorgriffin-handoff/install, then install with: npx skills add ConnorGriffin/skills --skill handoff

Registry 元数据

用于自动选择 Skill 的 Agent 可读档案。

本页通过 Registry API 提供相同的决策、信任、审计、场景和安装信号,让 Agent 无需抓取界面即可排序。

打开 Manifest

适配 Agent

58/100

GitHub automation

平台

Claude Code

审计报告

需审查 · 76/100

对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。

查看审计报告查看评估报告

Agent 决策面板

Fallback candidate for GitHub automation

先用此 Skill 做原型验证,并保留备选方案。

58
就绪度
原型验证
阶段

栈中角色

备选候选

主要匹配

GitHub automation

信任标签

先做原型验证

安装路径

命令已就绪

适用场景

  • GitHub automation 工作流
  • Claude Code 团队
  • builders willing to evaluate younger projects

证据

  • 仓库近期活跃
  • 已提供安装命令或 GitHub 仓库
  • 59/100 质量档案

先审查

  • Low GitHub adoption signal
  • Repository license is NOASSERTION; unclear licensing may hinder adoption.
  • 暂未有 OpenAgentSkill 使用反馈数据

实施路径

  1. 1在沙盒 Agent 中安装它,并端到端完成一次GitHub automation任务。
  2. 2Compare output quality, latency, and failure behavior against at least one alternative.
  3. 3Promote it into production only after reviewing repository permissions, license, and maintenance signals.

信任档案

仅限沙盒

有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。

64
OpenAgentSkill 信任评分

GitHub 采用度

修复

18 个 GitHub Stars

Star/Fork 活跃度

修复

18 个 Star,5 个 Fork; 当前元数据中没有议题活跃度信息

近期维护

通过

距上次推送 4 天

许可证清晰度

通过

NOASSERTION

积极信号

  • AI 审查已通过
  • 安装路径可用
  • 仓库证据可用
  • 近期维护的仓库
  • 安装命令未发现明显高风险模式
  • 结果闭环已就绪,但需要首次真实 Agent 运行

安装前审查

  • Repository license is NOASSERTION; unclear licensing may hinder adoption.
  • Financial research output is not financial advice; require human review before any live investment decision.
  • Low GitHub adoption signal
  • Quality score needs review
  • GitHub adoption: 18 GitHub stars
  • Stars/forks activity: 18 stars, 5 forks; issue activity unavailable in current metadata
  • 暂未有真实 Agent 结果报告
  • 无人值守安装前需要人工审查

建议操作

仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。

质量档案

有潜力 适用于 Agent 工作流的候选

有用的候选项,但采用前应与替代方案比较。

59
GitHub Stars
18
新鲜度
4 天前
安装就绪
许可证
NOASSERTION
安装前审查: Low GitHub adoption signal · Repository license is NOASSERTION; unclear licensing may hinder adoption.

工作流匹配

在这些场景使用此 Skill

工作流匹配

加入完整工作流

替代方案短名单

安装前对比

可能适合该任务的相近 Skill。

对比全部

概览

--- name: handoff description: Produce a decision-safe handoff for a fresh agent. Use when a task changes hands, a session is ending midstream, a user asks for a handoff, or an agent needs to preserve a design/map/implementation decision without making the next agent rediscover it. ---

# Handoff

Treat a handoff as a bounded state transfer, not a project report. Give the next agent the current decision, the evidence that makes it true, the safe first action, and the boundaries that prevent duplicate or contradictory work.

## 1. Establish the transfer

Identify the recipient and the job they are taking over. State whether the handoff is for a fresh agent, a later session, or a human decision.

Record the user's latest intent in their words where possible. Separate it from earlier superseded requests.

Finish when the receiving job is explicit and no longer depends on conversation archaeology.

## 2. Re-ground volatile state

Verify facts that may have changed: active agents and claims, issue or PR state, branch/worktree status, uncommitted files, and the latest rendered or test result. Do not carry forward status from memory.

Capture only artifacts the recipient needs, with exact paths or canonical issue/PR links. Mark each as committed, uncommitted, remote, local-only, retired, or merely referenced. Do not present an unpushed link as available work.

Finish when each material claim has a source the recipient can inspect.

## 3. Preserve decisions and boundaries

Write the settled decisions first. For each one, name its owner or source of truth and the consequence for the next agent.

Then record unresolved decisions as choices, not as accidental assumptions. If an ordering, authorization, product meaning, or dependency is genuinely undecided, make the first action a question or a safe investigation; do not choose silently.

List rejected, retired, or already-completed work that must not be repeated. Include domain vocabulary corrections that change implementation meaning.

Finish when the receiving agent can distinguish locked work from proposed work and pending work.

## 4. Give one safe first action

End with the next action that advances the task without reopening settled ground. If it is blocked by a user choice, say exactly what must be decided and why. Do not tell the recipient to “review the context” or “continue work.”

Finish when a fresh agent can begin productively from the handoff alone.

## Handoff shape

Use this shape; omit empty sections rather than adding filler.

```md # Handoff — <recipient job>

## First action

<one safe, specific action; include the required user decision if blocked>

## Current decision

- <settled decision> — <source of truth>; <consequence>

## Evidence and artifacts

- [artifact or issue](https://example.com/artifact-or-issue) — <status and why it matters>

## Do not redo or assume

- <retired/completed/rejected work> - <unresolved choice and its options>

## Safety and repository state

- <relevant hazards, claims, dirty-worktree or authority limits> ```

## Delivering it

**The handoff is the artifact. Write it to a file and hand over the path — nothing else.**

A handoff is written for the agent taking over, not for the person who asked for it. Asking to hand the work off means they are done reading. Summarizing the handoff back to them defeats the point: they read the same content twice, in the worse of the two formats, at the moment they were trying to stop.

The delivering response is one line — the path, plus the single fact only they can act on if one exists (an unmerged PR, a pending authorization). No section recap, no highlights, no "the parts you'd want to know", no bulleted extract of the file. If something genuinely cannot wait for the next agent, it belongs *in* the first-action section, not in a parallel chat summary.

This holds when the handoff is long and the urge to orient them is strongest. Length in the file is fine — it is written for a cold reader. Length in the response is not.

## Quality gate

Before delivering, verify that the handoff names the first action, distinguishes fact from proposal, links every material artifact, records volatile state as checked, and tells the recipient what not to repeat. Remove chronological narration, speculative detail, and generic next steps.

Then check the response itself: if it restates anything the file already says, cut it back to the path.

技术详情

版本
1.0.0
许可证
NOASSERTION
最近更新
2026年8月19日
发布时间
2026年8月19日

决策摘要

备选候选

58
就绪
原型验证
阶段

仓库近期活跃

审计

安装审查

安装与采用审查

76
需审查
安全性
80/100
维护状态
100/100
安装
92/100
打开完整审计查看评估报告

Agent 验证证据

Agent 验证证据

来自解析、审查、安装和一次小范围运行后的结果报告。

0
已验证
Needs first agent run自动安装: 先审查最近: 未知
成功率
近期失败
结果
0
输出质量
失败
0
不相关
0
安装次数
0
风险拦截
0
需要配置
0
生产环境
0

暂时没有 Agent 结果数据。首次 Agent 执行可以通过 /api/agent/outcome 报告成功、需要设置、风险拦截、失败或不相关。

安装

加入 Agent 工作流

免费且开源. 在生产 Agent 中安装前请先审查报告。

增长闭环

分享工具包

X

为 handoff 准备的场景化草稿,可手动发布到 X。

策展说明
handoff: Produce a decision-safe handoff for a fresh agent. Use when a task changes hands, a session i...

18 stars

https://www.openagentskill.com/skills/connorgriffin-handoff?ref=x
打开 X 草稿
可选:带安装命令的回复
Listing + install path for handoff:
https://www.openagentskill.com/skills/connorgriffin-handoff?ref=x

Install: npx skills add ConnorGriffin/skills --skill handoff
打开回复草稿

收录来源

Registry 收录

可认领

此列表来自公开来源,维护者认领获批前不会标记为官方。

创作者
ConnorGriffin
收录方
OpenAgentSkill 社区索引

归属链接指向公开仓库或创作者主页。创作者可认领列表以更新所有权信号。

认领此 Skill

所有者认领

认领此 Skill 页面

这条 Registry 收录 列表归属于 ConnorGriffin,但尚未标记为官方。认领后可增加已验证所有者信号,使后续发布、安装和审计更新更值得信赖。

创作者外链工具包

将证据徽章加入你的 README

在开发者评估仓库的位置展示规范页面、当前信任与审计信号,以及真实的 Agent 验证证据。

[![Listed on OpenAgentSkill](https://www.openagentskill.com/api/badge/connorgriffin-handoff?metric=listed&label=Listed)](https://www.openagentskill.com/skills/connorgriffin-handoff)
[![OpenAgentSkill Trust](https://www.openagentskill.com/api/badge/connorgriffin-handoff?metric=trust&label=Trust)](https://www.openagentskill.com/skills/connorgriffin-handoff)
[![OpenAgentSkill Audit](https://www.openagentskill.com/api/badge/connorgriffin-handoff?metric=audit&label=Audit)](https://www.openagentskill.com/skills/connorgriffin-handoff/audit)
[![Agent Proven](https://www.openagentskill.com/api/badge/connorgriffin-handoff?metric=proven&label=Agent%20Proven)](https://www.openagentskill.com/skills/connorgriffin-handoff)

作者

C

ConnorGriffin

@connorgriffin

平台适配

健康信号

GitHub Stars
18
质量评分
32/100
最近 GitHub 推送
2026年8月19日
框架提示
未知
OpenAgentSkill 浏览量
0
复制安装命令
0
跳转点击
0

社区信号

告诉我们这个 Skill 是否对你的 Agent 工作流有帮助。汇总反馈会持续改善排序。

信任与安全

仅限沙盒

64
  • GitHub 采用度18 个 GitHub Stars修复
  • Star/Fork 活跃度18 个 Star,5 个 Fork; 当前元数据中没有议题活跃度信息修复
  • 近期维护距上次推送 4 天通过
  • 许可证清晰度NOASSERTION通过
  • README/SKILL.md 完整度元数据包含足够的用法与工作流上下文通过
  • 依赖与运行时风险公开元数据中未发现主要依赖风险提示通过