> ## Documentation Index > Fetch the complete documentation index at: https://docs.hops.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Core concepts > The building blocks of Hops: Spaces, agents, and shared context. Hops has a few key ideas. Once you get these, everything else clicks. ## Spaces A Space is like a project room. It's where your team has conversations, shares docs, pins important context, and works with agents, all in one place. You might create a Space for a product launch, a client account, or an ongoing team channel. Everything inside that Space becomes context your agents can use. Create as many Spaces as you need. One per project or workflow keeps things clean and makes your agents more focused. ## Agents Agents are the AI members of your team. They live inside your Spaces, respond when you tag them with `@`, and use your team's actual context to get work done. A Space can have **multiple agents**. Tag any of them in the same conversation and they each respond from their own specialization. **The default agent** is called **Hops**. It's available in every Space right away, no setup needed. Just type `@Hops` and ask it something. **Custom agents** are ones you create for specific roles. Each agent can be specialized with: * **Custom instructions** that shape how the agent behaves, what it prioritizes, and what it avoids * **Artifacts** like documents, tables, and other reference material the agent uses as context * **Custom connections** giving per-agent access to apps and data from 500+ integrations across productivity, developer tools, CRM, communication, analytics, and more A marketing agent that knows your brand guide. A sales agent that pulls from your CRM. A support agent that reads your help docs. You decide what each one does, what it connects to, and how it should behave. ## Shared context Here's what makes Hops different from a regular AI chat: your agents remember. Every conversation, every pinned doc, every connected tool adds to the shared context in a Space. The next time someone asks an agent a question, it already has the background. No one has to re-explain the project. Context comes from: * Conversations in your Spaces * Docs and files your team has shared * Tools you've connected (Slack, Drive, Salesforce, etc.) * Previous decisions and work ## Connected tools Hops connects to 500+ apps across productivity, developer tools, CRM, communication, analytics, design, databases, and more, powered by the Pipedream integration catalog. When you connect a tool, your agents can read from it. Instead of you copying and pasting information, your agents go find it. The **default Hops agent** and your **personal agent** use your workspace connections automatically. Any tool connected at the workspace level is available to them right away. **Custom agents** get their own connections. You choose which apps each custom agent can access, so a marketing agent connects to your CMS and analytics, while a sales agent connects to your CRM. Each one only sees the tools relevant to its role. ## Visible work Every time an agent does something, it shows you exactly what it used and what it did. If it needs to do something with real-world consequences, like sending an email or updating a record, it asks for your approval first. You're always in control. Agents move the work forward. You make the decisions. ## What's next Create your first agent in under five minutes. How Spaces organize your team's work. How agents work in Hops. Plug in the tools your team already uses. ## Related topics - [Founders & executives](/docs/teams/founders-executives.md)