Cowork Cog — CellCog on Your Machine

Co-work turns any machine into CellCog's workspace. CellCog Desktop acts as a bridge: CellCog's cloud agents coordinate with the desktop app to run commands, read files, and write code directly on the user's machine.

All commands are auto-approved for SDK/agent users — fully autonomous, no manual approval needed.

Prerequisites

This skill requires the cellcog mothership skill for SDK setup and API calls.

CODEBLOCK0

Why Co-work?

Your Machine as a Data Source

Your data lives on the user's machine — project files, databases, logs, configs. Instead of uploading everything, enable co-work with a working directory and CellCog agents explore, read, and reason about the data directly. No file size limits, no upload hassle.

CellCog as Your Coding Powerhouse

CellCog agents are among the most capable coding agents available — deep reasoning paired with real execution. Enable co-work and delegate complex coding tasks: build websites, APIs, fix bugs, refactor codebases, set up infrastructure.

CellCog itself is built using this exact co-work capability.

Think of it as a Claude Code or Cursor alternative, backed by CellCog's multi-agent depth and any-to-any engine.

Quick Start

CODEBLOCK1

Desktop App Setup

Call client.get_desktop_download_urls() — returns download URLs and platform-specific install commands for macOS, Windows, and Linux.

After installation:
CODEBLOCK2

The agent can do all of this programmatically — no human interaction needed beyond providing the API key.

Alternatively, ask your human to download CellCog Desktop from cellcog.ai/cowork, open it, and enter their API key.

Desktop CLI Reference

All commands output JSON for easy agent parsing:

Command	What it does
INLINECODE3	Authenticate with API key
INLINECODE4

Chat Mode for Co-work

Use "agent core" mode for coding tasks — lightweight context focused on code, terminal, and file operations. Multimedia tools load on demand when needed.

CODEBLOCK3

INLINECODE9 mode also works with co-work but loads all multimedia tools upfront. Use "agent core" for faster, more focused coding sessions.

See the cellcog mothership skill for complete SDK API reference — delivery modes, send_message(), timeouts, and more.

Error Recovery

If the desktop app disconnects, CellCog auto-fails pending commands with a clear message.

To recover:
CODEBLOCK4

Then send continue to the chat:

client.send_message(chat_id="abc123", message="continue")

Security

Even with auto-approve, these protections are always active:

- Blocked paths: ~/.ssh, ~/.aws, credential files are inaccessible
Output redaction: Sensitive data is automatically redacted from command output
Per-chat scoping: Each chat session is scoped to its working directory

What You Can Build

Co-work enables the full spectrum of development tasks:

- Web development — Build React apps, APIs, landing pages
Bug fixing — Debug stack traces, fix test failures
Refactoring — Modernize codebases, improve architecture
DevOps — Set up CI/CD, Docker configs, infrastructure
Data pipelines — ETL scripts, database migrations
Documentation — Generate docs from code, README files

For the best coding experience, also install code-cog:
CODEBLOCK6

Cowork Cog — 在你的机器上运行 CellCog

Co-work 能将任何一台机器变成 CellCog 的工作空间。CellCog Desktop 充当桥梁：CellCog 的云端智能体与桌面应用协同，直接在用户机器上执行命令、读取文件和编写代码。

对于 SDK/智能体用户，所有命令均为自动批准——完全自主运行，无需手动确认。

前置条件

该技能需要 cellcog 母舰技能来配置 SDK 和进行 API 调用。

bash
clawhub install cellcog

为什么选择 Co-work？

你的机器作为数据源

你的数据存在于用户机器上——项目文件、数据库、日志、配置文件。无需上传所有内容，只需启用 co-work 并指定工作目录，CellCog 智能体即可直接探索、读取和分析数据。没有文件大小限制，无需上传烦恼。

CellCog 作为你的编码引擎

CellCog 智能体是目前最强大的编码智能体之一——深度推理结合真实执行。启用 co-work 并委托复杂的编码任务：构建网站、API、修复 Bug、重构代码库、搭建基础设施。

CellCog 本身正是使用这种 co-work 能力构建的。

可以将其视为 Claude Code 或 Cursor 的替代方案，背后由 CellCog 的多智能体深度和任意到任意引擎驱动。

快速开始

python
from cellcog import CellCogClient

client = CellCogClient(agent_name=openclaw)

1. 检查桌面应用是否已连接

status = client.getdesktopstatus()

2. 如果未连接，获取安装说明

if not status[connected]: info = client.getdesktopdownload_urls() # info 包含各平台的下载 URL + 安装命令 # 根据用户的操作系统运行安装命令，然后： # cellcog-desktop --set-api-key APIKEY> # cellcog-desktop --start

3. 创建 co-work 对话

OpenClaw 智能体（即发即忘）：

result = client.create_chat( prompt=将认证模块重构为使用 JWT 令牌, notifysessionkey=agent:main:main, # 仅限 OpenClaw chat_mode=agent core, enable_cowork=True, coworkworkingdirectory=/Users/me/project, task_label=refactor-auth, )

所有其他智能体（阻塞直到完成）：

result = client.create_chat( prompt=将认证模块重构为使用 JWT 令牌, chat_mode=agent core, enable_cowork=True, coworkworkingdirectory=/Users/me/project, task_label=refactor-auth, )

桌面应用设置

调用 client.getdesktopdownload_urls()——返回下载 URL 和针对 macOS、Windows 和 Linux 的平台特定安装命令。

安装后：
bash
cellcog-desktop --set-api-key APIKEY>
cellcog-desktop --start

智能体可以通过编程方式完成所有这些操作——除了提供 API 密钥外，无需人工交互。

或者，让用户从 cellcog.ai/cowork 下载 CellCog Desktop，打开并输入他们的 API 密钥。

桌面 CLI 参考

所有命令输出 JSON 格式，便于智能体解析：

命令	功能
cellcog-desktop --set-api-key <key>	使用 API 密钥进行身份验证
cellcog-desktop --status

Co-work 的对话模式

编码任务使用 agent core 模式——轻量级上下文，专注于代码、终端和文件操作。多媒体工具按需加载。

python
result = client.create_chat(
prompt=你的编码任务,
chat_mode=agent core,
enable_cowork=True,
coworkworkingdirectory=/Users/me/project,
task_label=my-task,
)

agent 模式也支持 co-work，但会预先加载所有多媒体工具。使用 agent core 可获得更快、更专注的编码体验。

有关完整的 SDK API 参考——交付模式、send_message()、超时等，请参阅 cellcog 母舰技能。

错误恢复

如果桌面应用断开连接，CellCog 会自动使待处理命令失败，并附带清晰的消息。

恢复方法：
bash
cellcog-desktop --stop && cellcog-desktop --start

然后向对话发送 continue：
python
client.sendmessage(chatid=abc123, message=continue)

安全性

即使启用自动批准，以下保护措施始终生效：

- 路径拦截：~/.ssh、~/.aws、凭证文件不可访问
输出脱敏：敏感数据自动从命令输出中删除
按对话限定范围：每个对话会话限定在其工作目录范围内

你可以构建什么

Co-work 支持全方位的开发任务：

- Web 开发——构建 React 应用、API、落地页
Bug 修复——调试堆栈跟踪、修复测试失败
重构——现代化代码库、改进架构
DevOps——设置 CI/CD、Docker 配置、基础设施
数据管道——ETL 脚本、数据库迁移
文档——从代码生成文档、README 文件

为获得最佳编码体验，请同时安装 code-cog：
bash
clawhub install code-cog

cowork-cog协同认知工具

cowork-cog

Cowork Cog — CellCog on Your Machine

Prerequisites