菜单
Review cowork project outputs and sync the workspace-level README.md and log.md in cowork_projects/. Use when cowork agents have done work and the workspace index needs to be updated, or when the user says "sync cowork", "review cowork status", or "update cowork workspace".
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/allyoushawn/agent_skills --skill cowork-sync命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Generic skill for tasks that require planning, debugging, or iteration before they can be completed. Three modes — execute / optimization / orchestrate — selected by the request shape. Discriminator: does the task require planning? If the request is unambiguous and mechanical ("run this notebook with python3"), do NOT use this skill — direct execution is correct. If achieving the goal requires planning, recovery from failures, or iterating toward a target, this skill applies. Triggers: "use achieve-goal to X", "iterate to achieve metric Y", "execute this multi-milestone plan", or any goal that needs planning + recovery.
Dispatch a review task to Gemini CLI (large-context review — plans, docs, large file corpora) or Codex CLI (code review — diffs, PRs, implementation quality). Routes based on the nature of the artifact being reviewed. Use from achieve-goal, literature-survey skills, or any skill that needs a CLI-based review gate. This is a headless CLI review gate over a concrete artifact (code, diff, or doc corpus); for an interactive second opinion or a side-by-side model comparison via MCP chat, use consult-llm instead.
Composes a narrative "war story" history of how a line of work, a project, or a research direction evolved — what was tried, what was believed, what broke that belief, what changed, and where it's heading. Produces a mental-model document a newcomer (or future-self) can absorb. Use when the user asks for "the history of how X evolved", "a war story", "how did we get here", "the backstory/trajectory of this work", or wants the reasoning arc behind a series of decisions. Verifies factual and chronological claims against sources (NotebookLM if available) to avoid timing/inspiration contradictions, and ends with a CLI review gate.
基于 SOC 职业分类
正在显示 SKILL.md
| name | cowork-sync |
| description | Review cowork project outputs and sync the workspace-level README.md and log.md in cowork_projects/. Use when cowork agents have done work and the workspace index needs to be updated, or when the user says "sync cowork", "review cowork status", or "update cowork workspace". |
Updates the workspace-level index (cowork_projects/README.md and cowork_projects/log.md) by reading each cowork project's current state. These workspace-level files are Claude Code's responsibility — cowork agents must not touch them.
kb-update to promote cowork outputs to canonical KB| File | Owner |
|---|---|
cowork_projects/README.md | Claude Code — this skill updates it |
cowork_projects/log.md | Claude Code — this skill appends to it |
cowork_projects/CONVENTIONS.md | Claude Code — never touch |
<project-name>/README.md | Cowork agent — read-only from this skill's perspective |
<project-name>/log.md | Cowork agent — read-only from this skill's perspective |
List all subdirectories inside knowledge_base/projects/cowork_projects/. Ignore files at the workspace root (CONVENTIONS.md, README.md, log.md). Each subdirectory is a cowork project.
For each project folder found, read:
<project-name>/README.md — current state, findings, proposed KB destination, open questions<project-name>/log.md — append-only event timelineClassify each project into one of three states based on its README:
Compare each project's most recent log.md date against the last sync entry in cowork_projects/log.md. Only projects with new log entries since the last sync need to be reflected in the workspace log.
Rewrite the workspace README to reflect the current state of all projects. Use this structure:
---
title: Cowork Projects
summary: Sandbox workspace for Claude.ai web ("cowork") task outputs. Claude Code reviews and promotes content from here into the canonical KB.
topics: [cowork, workflow, sandbox]
status: active
updated: <today's date>
---
<!-- BEGIN KB_INDEX -->
# cowork_projects/
Sandbox workspace for Claude.ai web ("cowork") task outputs. Cowork writes here; Claude Code reviews and promotes to canonical KB. See `CONVENTIONS.md` for cowork operating rules.
## Key Topics
cowork, workflow, sandbox, conventions, kb-maintenance
## Files
| File | Summary | Topics | Status | Updated |
|------|---------|--------|--------|---------|
| CONVENTIONS.md | Authoritative operating conventions for cowork agents: boundary rules, per-task folder structure, README vs log.md conventions, and handoff protocol | cowork, workflow, sandbox, conventions, kb-maintenance | active | <date> |
| log.md | Append-only timeline | — | active | <date> |
## Subfolders
| Folder | Description | Key Topics |
|--------|-------------|------------|
<one row per project, summary drawn from project README>
<!-- END KB_INDEX -->
## Active projects
<list each active project: name, one-line status, last log entry date>
## Pending promotion
<list each pending_promotion project , , >
## Stale projects
For each project that has new activity since the last sync, append one line:
## [YYYY-MM-DD] update | <project-name>: <one-liner drawn from most recent project log entry>
If multiple projects updated, write one line per project. Always add a final sync line:
## [YYYY-MM-DD] sync | cowork-sync ran; <N> project(s) reviewed, <M> updated
Summarize:
kb-update promotion is recommended