# Nuclino MCP for AI Agents AI Agent Connect > Nuclino MCP lets your AI agent read and write to your Nuclino workspaces. It gives your agent a direct line to your company's knowledge base, project docs, and team structures. Use it to search for internal specs, create new wiki pages, or pull the latest project updates without leaving your chat. ## Overview - **Category:** productivity - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_Z6PHGHm46XMsfoT4XFEq1GBWsSKYDu8jtNxAZdoa/ai-agent-connect - **Tags:** team-wiki, document-collaboration, workspace-management, knowledge-base, project-tracking ## Description Nuclino MCP connects your AI agent to your company's unified workspace for knowledge, docs, and projects. Imagine your team's entire knowledge base is just a conversation away. Instead of hunting through folders or asking coworkers for the latest project brief, you can just ask your agent to find it. This connection lets your agent actually see your documents, project pages, and team structures. You can have your agent draft a new wiki page based on a meeting transcript, update an existing project status, or pull specific technical specs into a chat. It turns your company wiki from a static repository into a dynamic resource that your agent can interact with in real time. Because Vinkius hosts this Connector, it's easy to plug into your existing setup and get your agent up to speed on your organization's specific documentation. You don't have to copy and paste anything; the agent just knows where the info is and what it says. This is especially useful for keeping technical documentation current as your projects evolve, ensuring that everyone on the team is looking at the same source of truth. It removes the friction of manual updates and makes your internal wiki a living part of your daily workflow. By integrating this into your daily routine, you ensure that project requirements and team guidelines stay accessible to everyone, regardless of where they are working. ## Tools ### list_teams Shows all teams you belong to so the agent knows where to look. This helps the agent navigate your organization's root structure. ### list_workspaces Lists the workspaces inside a team to help the agent narrow down its search. It provides the necessary IDs for scoping queries. ### list_items Shows all the pages and documents inside a specific workspace. Use this to see what content is currently available. ### get_item Pulls the full content and configuration of a specific document. This allows your agent to read the exact Markdown payload. ### create_item Lets the agent write a brand new wiki page or document into a workspace. This creates permanent documentation from your prompts. ### update_item Overwrites or appends new information to an existing document. It allows for real-time edits to your wiki content. ### delete_item Permanently removes a document from the workspace. This should be used with caution to manage your knowledge base. ### search_items Performs a global search across a team to find specific documents. It uses indexed search to uncover unknown document IDs. ### list_collections Shows how documents are grouped into folders or directories. This helps the agent understand the document relationship graph. ### list_fields Maps out the custom fields and properties used in your workspace. This helps the agent understand your structured data dimensions. ### list_users Lists the people who are part of your team. This provides context on who owns different parts of the workspace. ### list_files Shows the actual file attachments linked to a document. This exposes the URL bindings for binary data records. ## Prompt Examples **Prompt:** ``` Search Nuclino for any documentation mentioning 'SSO Security Policies'. ``` **Response:** ``` I've successfully performed a fuzzy text search and located the page **SSO Security Implementation Guidelines**. I can extract its contents or share the browser link if needed. Would you like me to summarize the key requirements for you? ``` **Prompt:** ``` Create an item titled 'Project X Architecture Brief' in the Engineering workspace. ``` **Response:** ``` I retrieved the Engineering workspace ID and created the new item **Project X Architecture Brief**. The document is live and awaiting further edits. Let me know if you'd like me to add any specific technical details to it now. ``` **Prompt:** ``` List all teams connected to this authentication token. ``` **Response:** ``` You have access to 3 teams on this environment: * **Design Ops** (`id-1`) * **Software Engineering** (`id-2`) * **Executive Planning** (`id-3`) Which team would you like to explore first? ``` ## Capabilities ### Search the entire knowledge base Your agent can find specific documents or notes across your whole team's workspace instantly. ### Create new wiki pages Have your agent draft and publish new documentation based on your verbal instructions. ### Update existing documentation Modify current project specs or meeting notes without leaving your AI chat interface. ### Retrieve document content Get the full text and configuration of any page in your workspace for quick reference. ### Map team hierarchies Keep your agent aware of your team structure, workspaces, and member lists. ## Use Cases ### Finding security policies An engineer asks for the SSO policy and the agent finds the specific document in the security workspace using a global search. ### Creating a project brief A PM asks the agent to create a new page for 'Project X' in the Engineering workspace to capture initial requirements. ### Updating meeting notes A manager tells the agent to append today's action items to the 'Weekly Sync' page to keep the record current. ### Auditing workspace structure A knowledge manager asks the agent to show how folders are organized to identify areas that need better categorization. ## Benefits - Stop wasting time searching for documents. Use your agent to find specific project briefs or internal notes instantly across your entire team. - Create documentation faster. Turn meeting transcripts or rough notes into full wiki pages with a single natural language command. - Keep project specs current. Update existing documents and append new information without having to open a browser tab or manual editor. - Get context on your team. Help your agent understand your organization's hierarchy by letting it see your teams, workspaces, and members. - Access files directly. Let your agent see which attachments are linked to your documents so you can quickly reference external resources. ## How It Works The bottom line is your AI agent gets a direct window into your company's internal documentation. 1. Subscribe to the Nuclino MCP on Vinkius. 2. Enter your Nuclino Personal API Key to authorize the connection. 3. Ask your agent to find, create, or update documents in your workspace. ## Frequently Asked Questions **Can the Nuclino MCP help my agent find internal docs?** Yes. It lets your agent perform a global search across your entire team's workspace to find specific pages or notes instantly. **Can I create new wiki pages using the Nuclino MCP?** Yes. Your agent can write new documents directly into your workspace using natural language commands, making it easy to turn notes into permanent wiki pages. **Does the Nuclino MCP support updating existing content?** Yes. You can tell your agent to append notes or overwrite sections of a page, and it will update the live document for the whole team. **Can my agent see file attachments in Nuclino?** Yes. It can list the files and URLs attached to any item, so your agent knows what extra resources are available for a specific document. **Can my agent see my whole team structure?** Yes. It can list your teams, workspaces, and collections to understand how your company's knowledge is organized and who is involved. **Is the Nuclino MCP good for project management?** It's great for managing project documentation. You can use it to create planning pages, update status notes, and search for project requirements quickly. **How can I explore the hierarchy of my company's Nuclino configuration?** Your AI agent can progressively drill down by invoking `list_teams`, taking those IDs into `list_workspaces`, checking the clusters via `list_collections`, and finally dumping the granular content using `list_items`. **Can the agent perform global searches if I don't know the workspace?** Yes. Instead of manually parsing directories, ask the agent to invoke the `search_items` tool. It queries all permissible areas on Nuclino simultaneously and returns exact contextual matches within seconds. **Can I automatically append meeting notes as a brand new document?** Absolutely. Once an AI process finishes an important chat, use the `create_item` tool to generate a fresh target URL holding the transcribed content straight into any specified workspace.