new-convo-handoff SKILL.md skillA

new-convo-handoff from sichenai/sichen-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add sichenai/sichen-skills/new-convo-handoff@git:20260828.a14f647

Pins git:20260828.a14f647 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
sichenai/sichen-skills
Path
skills/new-convo-handoff/SKILL.md
Versions recorded
3
Latest version
git:20260828.a14f647, committed Aug 28, 2026
First version recorded
Aug 8, 2026
Size
18,311 bytes
Repository stars
3
License
MIT

新对话上下文交接指令生成器(v2.0,支持跨体模式)。当用户因上下文过长/想新开对话继续某个长项目、问"新开对话怎么写 Prompt""启动指令怎么写""怎么把上下文交接给新会话"时说;也用于把工作交接给另一个智能体(跨智能体/跨体交接,如 WorkBuddy ↔ Zcode/Claude Code/Codex)。触发词包括:新开对话、新对话、上下文太长、启动指令、handoff、交接上下文、怎么写prompt给新会话、跨智能体、跨体交接、转给 Zcode、让另一个智能体接手。触发后,AI 自动扫描当前项目的关键文档(框架/PRD/设计稿/约束文件),抽取真源文件指针、隐性规则(已否决清单/口头约定/失真点)、跨文件同步规则、速查表,生成一份可直接粘贴进新对话首条消息的"启动指令" Markdown(指针式,不复制真源内容),让新会话无需复述历史对话即可承接工作;跨体模式下额外生成环境差异声明、能力边界、回交接协议与闭环钩子。基于 2026-08-27 WorkBuddy↔Zcode POC 实测验证。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# 新对话上下文交接指令生成器

## 何时触发

用户出现以下任意信号时启用本 skill:
- 明确说"新开对话""新对话""上下文太长/太满,开个新会话"
- 问"新开对话怎么写 Prompt""启动指令怎么写""怎么把上下文交接给新会话"
- 提到 handoff / 上下文交接 / 让另一个会话接着干
- **跨体信号(v2.0 新增)**:提到"转给 Zcode/Claude Code/Codex/另一个智能体""跨智能体/跨体交接""让 XX 工具接手这个项目"

目标:产出一份**指针式**交接指令文档(指针 + 隐性规则 + 时效自检,不复制真源内容)。**两种模式**:
- **同体模式**:粘贴到同一智能体的新对话首条消息,按指针承接工作。
- **跨体模式**:交接给另一个智能体(如 WorkBuddy ↔ Zcode)。目标智能体读指令后不仅承接任务,还继承纪律(红线/否决清单),并把执行结果按协议写回执,形成闭环。

**模式判定(Step 0)**:出现跨体信号或用户指定了目标智能体 → 跨体模式;否则同体模式。判定不了就问一句。

## 跨体模式的三条铁律(2026-08-27 POC 实测教训,违反即翻车)

1. **回执闭环必须有钩子**:光写"对方写回执"是幽灵闭环——发起方智能体没有任何机制会自动读回执。**必须在发起方的项目记忆文件(MEMORY.md 或等价物)写入钩子**:「若 outputs/ 下有 `xx_回执_*.md`,任何会话先读最新回执再继续相关工作」。POC 首版漏此钩子被对抗性审查抓为严重问题(幽灵闭环)。
2. **红线用指针不用复制,但必须设硬前提**:安全红线(不编造/来源声明/溯源纪律)若半复制半指针,两头不靠——对方跳读就漏。正确做法:指令顶部设「硬前提」节(执行任何任务前先读记忆文件全文),红线节只写一行指针。理由:复制=第二真相源会漂移,但对方环境无自动注入,必须有强制读取动作兜底。
3. **自相矛盾自检**:生成后通读全文,查「修改/同步规则」与「禁止编辑」是否打架(POC 首版让 Zcode 写 `.workbuddy/memory/`,同时声明该目录只读——修复时才抓到)。记忆写入一律收归发起方:对方只写回执,不写记忆。

## 核心原则

- **指针优先,禁止复制(2026-08-04 第一性原理重构)**:启动指令 = 指针 + 防破坏清单 + 时效自检。真源文件里已有的约束细节(色值、字号、功能清单、优先级等)**一律不复制进指令,只给文件名和章节号**——复制即产生第二真相源,产出即开始漂移。
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add sichenai/sichen-skills/new-convo-handoff@sha256:c60284aaa7d1a6e9

The line at the top of this page pins the label git:20260828.a14f647, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_gqqwfpkg6seln7gf.svg)](https://markdownregistry.com/a/art_gqqwfpkg6seln7gf)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 3 versions of this file, between Aug 8, 2026 and Aug 28, 2026, about one every 10 days.

versioncommittedcommitsizeaudit
git:20260828.a14f647 latest2026-08-28 a14f647 18,311 BA view · diff
git:20260827.2e062332026-08-27 2e06233 17,390 BA view · diff
v1.0.02026-08-08 59683d0 10,794 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (18311 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

sichenai/sichen-skills · 3 stars · license MIT · pushed 2026-10-27 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_gqqwfpkg6seln7gf
GET https://markdownregistry.com/api/v1/resolve?ref=sichenai/sichen-skills/new-convo-handoff
GET https://markdownregistry.com/api/v1/blob/c60284aaa7d1a6e9bcf928e0c05f7fe5ba5fd6def1bdde4df415b76a032be4fc

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from sichenai/sichen-skills

adversarial-review SKILL.md skill
sichenai/sichen-skills · skills/adversarial-review/SKILL.md · 以攻击者视角审查 AI 交付的方案/设计/代码(AI 审 AI)。当用户说"对抗性审查""验收一下""这个方案靠谱吗""AI 写的这个能信吗""帮我挑刺""adversarial review""红队审查",或在编程/写作 agent…
v1.1.1 · audit A · 3 stars
browser-login-reuse SKILL.md skill
sichenai/sichen-skills · skills/browser-login-reuse/SKILL.md · 浏览器自动化登录态复用。当用户需要让 AI 操作真实浏览器完成需要登录的网站任务(如登录控制台查额度、提交表单、上传文件、面板操作),且希望"登录一次、之后复用登录态、不反复登录"时使用。触发词包括:浏览器自动化、登录态复用、帮我登录、操作…
v2.1.0 · audit B · 3 stars
clarify-until-clear SKILL.md skill
sichenai/sichen-skills · skills/clarify-until-clear/SKILL.md · 反复澄清确认模式。当用户说"反复确认""反复提问""确认意图""澄清意图""直到我确认""先别做,先问"或明确要求 AI 不要急于执行、先通过连续提问确认真实想法时触发。触发后,AI…
v1.0.0 · audit A · 3 stars
first-principles SKILL.md skill
sichenai/sichen-skills · skills/first-principles/SKILL.md · 第一性原理分析模式。当用户说"第一性原理""第一性原则""从第一性原理出发""first principles"或明确要求剥离假设、回到最基本事实重新推理时触发。触发后,AI…
v1.0.0 · audit A · 3 stars
model-connector SKILL.md skill
sichenai/sichen-skills · skills/model-connector/SKILL.md · 自定义大模型自动接入工程师(宿主无关版,仅覆盖 OpenAI 兼容及厂商双协议端点)。当用户说"接入自定义大模型""配置自己的大模型""把模型接入本Agent""接入工程师""接一个自己的模型""接 XX 模型",或希望把第三方/自建…
v1.21.0 · audit A · 3 stars

Every file in sichenai/sichen-skills

Browse by kind, by grade A, or by owner.