Workspace Overview

The workspace is the most fundamental organizational unit of Nekro Agent's capabilities.

Feature List

Feature	What It Does
Claude Code Sandbox	Execute code, process files, generate reports, complete complex tasks
Knowledge Base	Upload documents and materials so the AI can answer questions based on your resources
Memory System	Record long-term context, preferences, and project status to reduce repetitive explanations
Resource Center	Manage servers, databases, tokens, and other resources, mounting them to workspaces as needed
MCP Services	Connect external tools and services to extend workspace capabilities
Workspace Timers	Scheduled reminders, periodic inspections, and automated execution of recurring tasks
Prompts & Skills	Set long-term requirements and specialized capabilities for the workspace
Communications & Extensions	View associated records, manage extension capabilities and workspace configurations

Recommended Workflow

We recommend reading and configuring in the order of Deployment -> Plugins -> Workspace Overview: complete basic deployment first, then understand plugin capabilities, and finally create a workspace to host long-term tasks.

When you need the AI to do more than just chat — to actually help you get things done — organizing materials, analyzing data, writing reports, tracking projects, performing periodic inspections — the workspace is where all of this happens.

A workspace brings together the AI execution environment (Claude Code Sandbox), knowledge base, long-term memory, external tools (MCP), and scheduled tasks around a shared objective, working continuously toward the same goal. Conversations in channels are no longer one-off Q&A sessions but an ongoing project space.

What Workspaces Bring

From "Chatting" to "Getting Things Done"

Without a workspace, the AI can only answer questions in conversation. With a workspace, the AI can:

• Write and execute code in the Claude Code Sandbox, process files, and generate reports
• Connect to external tools and data sources through MCP services
• Automatically remember what was discussed and what was done before
• Automatically execute periodic tasks according to your schedule

Isolated Task Spaces

Each workspace is independent. The knowledge, memory, and tool configurations of different projects do not interfere with each other. You can run a "Data Analysis" workspace and a "Content Creation" workspace simultaneously, each maintaining its own context and capabilities.

Gets Better Over Time

The memory system in the workspace continuously extracts key information from conversations and tasks. The longer you use it, the deeper the AI understands your project, and the less frequently you need to re-explain context.

Workspace Capability Components

Capability	Function	Depends on CC Sandbox
Claude Code Sandbox	Core engine for code execution, file processing, and complex tasks	—
Knowledge Base	Upload documents; AI automatically retrieves and references them	No
Memory System	Automatically extracts and retains long-term context; gets smarter with use	No
Resource Center	Manage reusable servers, databases, tokens, and other resources; mount to workspaces as needed	Required for Claude Code usage
Timers	Let AI execute tasks on schedule	No
CC Model Group	Independently configure models for the Claude Code Sandbox	—
Prompts & Skills	Workspace-level behavior customization and capability configuration	No

Recommendation

We recommend creating an independent workspace for each actively used channel. Even if you don't need the Claude Code Sandbox right away, the knowledge base and memory system can significantly improve AI performance.

Workspace Detail Page

After entering a workspace, you can manage its capabilities through different tabs:

Tab	Purpose
Overview	View basic workspace status and main capability entry points
Sandbox	Manage the Claude Code Sandbox and task execution environment
Communications	View associated channels, conversations, and task records
Memory	Manage workspace long-term memory
Knowledge Base	Upload, bind, and search workspace materials
Extensions	Manage extension capabilities related to the workspace
MCP	Configure MCP services available to the workspace
Resources	Mount servers, databases, tokens, and other resources
Prompts	Set long-term requirements and behavior preferences for the workspace
Configuration	Adjust basic workspace parameters and runtime options

When using for the first time, you don't need to configure all tabs at once. We recommend starting with basic information, knowledge base, or memory, then enabling the sandbox, MCP, resources, and skills as needed for your tasks.

Typical Use Cases

Project Management Assistant

Create a workspace, upload project documents to the knowledge base, and bind a project discussion channel. The AI will remember project progress, and you can ask it to periodically generate progress reports and track to-do items.

Data Analysis Workbench

Create a workspace and enable the Claude Code Sandbox. Simply send data files to the AI, and it can analyze data with Python in the sandbox, generate charts, and send the result files back to you.

Automated Operations

Create a workspace, configure MCP services to connect to your monitoring system, and set up a timer to check every hour. The AI will automatically perform inspections and notify you when anomalies are detected.

Knowledge Management Center

Create a workspace and continuously upload team documents to the knowledge base. Any member asking questions in the channel will receive accurate answers from the AI based on the knowledge base.

Long-Term Companion

Create a workspace and enable the memory system. The AI will remember your conversations, preferences, and habits, providing a chat experience far beyond ordinary conversations.

Getting Started

Before using workspaces, we recommend completing the prerequisites:

1. Quick Deployment — Get Nekro Agent up and running first
2. Plugin Documentation — Understand what plugins can extend, and enable the necessary ones first
3. Workspace Overview — Then create a workspace to host long-term tasks

For more on workspace features, continue reading:

1. Quick Start — Build your first workspace in five steps
2. Claude Code Sandbox — Unleash the AI's execution capabilities
3. Knowledge Base — Let the AI master your professional materials
4. Memory System — Let the AI get to know you better over time
5. MCP Service Management — Connect external tools
6. Resource Center — Manage and mount reusable credentials and connection information
7. Workspace Timers — Implement automated workflows
8. Skills Library — Add specialized capabilities to Claude Code