Wolai MCP. Manage pages, blocks, and databases via conversation.

Wolai lets your AI client treat a massive knowledge base like an extension of its own brain. You don't have to click through menus or navigate complex file structures; you just ask for the data, and your agent gets it. It handles everything from simple note organization to structured product roadmaps.

It’s built around three core concepts: Pages (the main documents), Blocks (individual pieces of text or media inside those documents), and Databases (structured tables). Your AI client runs a series of tool calls—like checking the page list, then retrieving metadata, followed by running a targeted query—to build an answer for you.

It keeps the entire system organized so your agent acts like a real-time knowledge expert.

Managing Content and Pages

Need to know what documentation exists? You run list_pages to generate a comprehensive inventory of every accessible page in the workspace. If you need more detail on a specific document, you use get_page; this fetches all the metadata and content details for that single page ID. When you're ready to write something new, simply call create_page, and your agent generates an entirely blank slate inside Wolai.

To dive deep into what’s actually in a document—the individual text snippets or images—you list all the content blocks using list_blocks on any given page. You can also get a high-level overview of the entire environment by running get_workspace_info, which gives you context about the whole Wolai setup.

Working with Structured Data and Databases

This is where your agent shines. When documentation gets too messy for simple text, you use databases. To see what structured data sources are available, call list_databases to get an inventory of every database in the workspace. Before running any query, it's smart to check the structure itself; get_database retrieves the schema and definition for a specific table, so your agent knows exactly how to talk to it.

When you need data—say, all sales reps who worked on Project X last quarter—you run a filtered search using query_database. This function runs targeted queries against a database, pulling back only the rows that match your exact criteria. If your agent needs to add new information, it uses create_database_row to insert a brand-new, structured row of data into an existing table.

Team Access and System Oversight

Your AI client doesn't just read; it manages people too. You can pull up the current user roster in the workspace by calling list_users, which returns an array containing every user account associated with the platform. This helps you manage who has access to what knowledge.

The tools available are:

• create_database_row: Adds a new, structured row of data into an existing database table.
• create_page: Creates an entirely new page within the Wolai workspace.
• get_database: Retrieves the schema and structure definition for a specific database table.
• get_page: Fetches all metadata and content details for a single, specified page ID.
• get_workspace_info: Gathers high-level information about the entire Wolai workspace environment.
• list_blocks: Lists all content blocks (text, images, etc.) that exist within a specific page.
• list_databases: Provides an inventory of every database available in the workspace.
• list_pages: Generates a list of all accessible pages across the entire workspace.
• list_users: Returns an array containing all user accounts currently associated with the workspace.
• query_database: Runs a filtered query against a database to retrieve specific rows based on criteria.