Token导航 LogoToken导航

Agent Skill · 研究检索

generate-handoff-document(生成交接文档)

generate-handoff-document 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

generate-handoff-document 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,674 个 Skill 里排第 53,604。

总安装 528周安装 22GitHub Stars 公开资料未说明下载 176核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:generate-handoff-document(生成交接文档)
来源仓库:https://github.com/b-mendoza/agent-skills
仓库路径:skills/generate-handoff-document
安装命令:
npx skills add https://github.com/b-mendoza/agent-skills --skill generate-handoff-document
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/b-mendoza/agent-skills --skill generate-handoff-document

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

generate-handoff-document 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。

安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。当前顶部介绍为空,原始 SKILL.md 摘录未提供。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/generate-handoff-document

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/b-mendoza/agent-skills --skill generate-handof…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

generate-handoff-document 怎么安装?

在终端执行 npx skills add https://github.com/b-mendoza/agent-skills --skill generate-handoff-document。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

generate-handoff-document 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

generate-handoff-document 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

generate-handoff-document 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

generate-handoff-document 用的人多吗?

公开数据是累计安装 528 次,近一周安装 22 次,下载 176 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

generate-handoff-document 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
generate-handoff-document

生成交接文档 · 本页

528—只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Generating Handoff Documents

Create a resumable handoff package for an in-progress analytical session. This orchestrator does three things: think (identify missing inputs and failed gates), decide (choose the next specialist or targeted rerun), and dispatch (send extraction, validation, and assembly work to co-located subagents). The working data lives on disk as structured artifacts; the orchestrator keeps only verdicts, paths, counts, and unresolved questions in context.

Inputs

InputRequiredExample
TARGET_FILEYesdocs/auth-review-handoff.md
SUBJECTNoAuthentication review
TRACKING_FILESNodocs/auth-review-notes.md,docs/plan.md
CONTEXT_SOURCENocurrent conversation or docs/transcript.md

If the user omits optional values, infer them from the session when that is safe. Read ./references/data-contracts.md when deriving sibling artifact paths or checking the structured artifact schemas.

Workflow Overview

  1. context-extractor writes the instruction/Q&A artifact.
  2. insight-documenter writes the insights artifact.
  3. claim-validator optionally writes the claims artifact when tracking files exist.
  4. document-assembler reads those artifacts and writes the final handoff document.
  5. The orchestrator validates the final document and reruns only the failing stage(s).

Subagent Registry

Read a subagent definition only when you are about to dispatch it.

SubagentPathPurpose
context-extractor./subagents/context-extractor.mdCapture the original mandate, instruction amendments, and chronological Q&A history
insight-documenter./subagents/insight-documenter.mdExtract evidence-backed findings, risks, and recommendations from the session
claim-validator./subagents/claim-validator.mdVerify factual claims from external tracking files against primary sources when available
document-assembler./subagents/document-assembler.mdAssemble the final handoff document from the structured artifacts

How This Skill Works

  • Read ./references/data-contracts.md when you need the sibling artifact naming rules, JSON schemas, or final-section requirements.
  • Pass only explicit handoffs between stages: source path, target artifact path, subject, and optional tracking files.
  • Keep only verdicts, file paths, counts, and actionable warnings from each subagent. The detailed payload stays in the artifact files.
  • Treat claims from tracking files as provisional even after validation. The final handoff must preserve that caution for the next agent.
  • Use targeted fix cycles. If source extraction is incomplete, rerun the relevant upstream subagent and the downstream stages that consume it. If the problem is only document coherence or formatting, rerun document-assembler only.
  • Stop and surface the blocker if dispatch is unavailable or if three fix cycles fail to produce a coherent handoff.

Output Contract

All files written by this skill are resumability artifacts. They preserve workflow state for later continuation; they are not product-code changes.

ArtifactProduced byPurpose
TARGET_FILEdocument-assemblerFinal cold-start handoff document
<stem>.context.jsoncontext-extractorOriginal instructions, amendments, and Q&A log
<stem>.insights.jsoninsight-documenterFindings with evidence, category, priority, and verification state
<stem>.claims.jsonclaim-validatorOptional claim-validation checklist and summary

On success, TARGET_FILE contains five sections:

  1. Original Instructions & Scope
  2. Q&A Log
  3. Observations & Insights
  4. Unverified Claims & Validation Checklist
  5. Open Questions & Recommended Next Steps

If no tracking files were provided, Section 4 must explicitly say so and tell the next agent to verify any factual claims independently.

Phase Guide

SituationReference filePurpose
Contract or artifact-path questions./references/data-contracts.mdDerive sibling artifact names and confirm the schemas/required sections

Execution Steps

  1. Confirm TARGET_FILE and derive the sibling artifact paths described in ./references/data-contracts.md.
  2. Dispatch context-extractor with CONTEXT_SOURCE and CONTEXT_FILE.
  3. Dispatch insight-documenter with CONTEXT_SOURCE and INSIGHTS_FILE.
  4. If TRACKING_FILES exist, dispatch claim-validator with TRACKING_FILES, INSIGHTS_FILE, and CLAIMS_FILE. Otherwise record the claims step as skipped.
  5. Dispatch document-assembler with TARGET_FILE, SUBJECT, CONTEXT_FILE, INSIGHTS_FILE, and optional CLAIMS_FILE.
  6. Validate the written handoff against this checklist:

- every required section exists - each insight has rationale plus evidence - the claims section includes either the validation directive or the explicit "no tracking files" note - open questions are listed or explicitly marked resolved - a fresh agent could continue without consulting prior chat history

  1. If a check fails, rerun only the failing stage(s); cap the fix loop at three passes.
  2. Return the final handoff path plus a concise summary of counts, warnings, and open questions.

Example

Derived working artifacts:

  • docs/auth-review-handoff.context.json
  • docs/auth-review-handoff.insights.json
  • docs/auth-review-handoff.claims.json
  1. Dispatch context-extractor
  2. Subagent returns: CONTEXT: PASS File: docs/auth-review-handoff.context.json Q&A exchanges: 4
  3. Dispatch insight-documenter
  4. Subagent returns: INSIGHTS: PASS File: docs/auth-review-handoff.insights.json Insights: 6
  5. Dispatch claim-validator
  6. Subagent returns: CLAIMS: WARN File: docs/auth-review-handoff.claims.json Claims checked: 9 Unverified: 2
  7. Dispatch document-assembler
  8. Subagent returns: HANDOFF: PASS File: docs/auth-review-handoff.md Open questions: 2
  9. Report: "Handoff document written to docs/auth-review-handoff.md. The session is resumable from disk; two open questions remain."

The orchestrator keeps only those summaries and file paths.

平台分布

Codex

36.1%
按下载量换算64

Claude

30.93%
按下载量换算54

Cursor

17.64%
按下载量换算31

Gemini CLI

9.47%
按下载量换算17

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills