GitBook MCP. Query and manage your entire knowledge base via AI chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
GitBook MCP Server gives your AI agent full control over your technical documentation. List organizations, audit spaces, read full document pages, and run cross-page searches across your entire knowledge base.
You can manage docs-as-code workflows and retrieve technical information directly from any AI client.
What your AI agents can do
Get me
Retrieves the authenticated user's profile information.
Get page
Gets the full content of a specific document page.
Get space
Retrieves detailed metadata for a specific documentation space.
List organizations and spaces to map out your entire documentation hierarchy.
Retrieve the full text and metadata for any single page in your knowledge base.
Execute cross-page searches within your GitBook namespaces using natural language queries.
List and retrieve detailed metadata about collections and spaces to verify structure and access rules.
Extract your authenticated user's profile metadata (name, email) to confirm the agent's current permissions.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
GitBook MCP Server: 8 Tools for Documentation Management
Use these tools to programmatically list, search, and retrieve structured data from your GitBook documentation.
019d75a5get me
Retrieves the authenticated user's profile information.
019d75a5get page
Gets the full content of a specific document page.
019d75a5get space
Retrieves detailed metadata for a specific documentation space.
019d75a5list collections
Lists all content collections within your organization.
019d75a5list organizations
Retrieves a list of all connected GitBook organizations.
019d75a5list pages
Lists all individual pages contained within a specific space.
019d75a5list spaces
Lists all documentation spaces within a given organization.
019d75a5search content
Searches for content snippets across your entire GitBook documentation.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with GitBook, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
GitBook MCP Server - Manage Docs and Knowledge
Your AI agent gets full control over all your technical documentation. You can list organizations, audit spaces, read full document pages, and run cross-page searches across your whole knowledge base. It lets you manage docs-as-code workflows and pull technical info straight from any AI client.
Discovering the Docs Structure
To map out your entire documentation hierarchy, you can first use list_organizations to pull a list of all connected GitBook organizations. From there, list_spaces lets you pull all the documentation spaces inside a specific organization. Then, list_pages pulls every individual page inside a given space. To get a comprehensive list of content collections, use list_collections.
Reading Specific Content
Need to read a whole document? get_page pulls the full content of any specific page. You can also get detailed metadata about a specific space using get_space.
Searching Your Knowledge Base
Want to find an answer buried deep in your docs? search_content runs cross-page searches across your entire GitBook documentation using natural language queries. You can also search for content snippets across your whole documentation.
Auditing and Context
Check the structure and access rules for your docs by listing all spaces with list_spaces and pulling detailed metadata using get_space. You can verify the agent's current permissions and context by using get_me to retrieve the authenticated user's profile information, which pulls your name and email.
Putting It Together
Your agent can list all organizations with list_organizations. You can then map out the docs by calling list_spaces on an organization, and then list_pages on a space. If you need to check the details of a space, get_space pulls that metadata. When you're done mapping, and you need to find something specific, you just fire up search_content with what you're looking for.
To read a whole page, you use get_page. You'll always have the ability to confirm who the agent is working for using get_me.
How GitBook MCP Works
- 1 Subscribe to the GitBook server and input your GitBook API Token.
- 2 Your AI client initiates a request (e.g., 'List all spaces in the 'API Reference' org').
- 3 The server executes the necessary tool calls, pulls the structured data, and returns the findings to your agent for a conversational response.
The bottom line is, your AI agent treats your GitBook knowledge base like a native, queryable API, letting you manage docs without opening the web UI.
Who Is GitBook MCP For?
Technical Writers, Product Managers, and Support Engineers use this server. If your job involves knowing where a specific technical answer lives, or if you spend time auditing documentation structure, this is for you. You need to move beyond manually clicking through dozens of pages to querying the entire knowledge base with plain language.
Verifies content structure across multiple spaces and searches for existing documentation snippets without having to open the GitBook UI.
Audits documentation collections and spaces to understand how product knowledge is distributed across different parts of the organization.
Quickly searches for technical answers across all product spaces in real-time to resolve customer inquiries.
What Changes When You Connect
- Audit documentation spaces and collections using
list_spacesandlist_collections. You see exactly how product knowledge is grouped and where gaps exist, without needing to navigate the site. - Stop guessing where an answer is. Use
search_contentto run cross-page searches across all namespaces, pulling relevant snippets instantly based on natural language queries. - Verify page content instantly.
get_pagelets your agent retrieve the full text of a document, perfect for copying data or summarizing technical specs. - Map your entire doc structure. Running
list_organizationsandlist_spaceslets you build a complete map of all documentation environments in one go. - Check permissions and context.
get_mepulls your user profile metadata, letting you confirm the agent's current access limits before running complex tasks. - Analyze structure metadata. Use
get_spaceto fetch detailed information about a specific documentation space, verifying its visibility and structural rules.
Real-World Use Cases
Need to find the API flow for a new feature.
A developer needs to confirm the 'authentication flow' but doesn't know which space it's in. They ask their agent, 'Search for authentication flow.' The agent uses search_content and finds the most relevant page in the 'API Reference' space, giving the developer the exact ID and title.
A new PM needs to audit product knowledge.
A Product Manager needs to know if the 'Billing' documentation is properly separated from the 'User Guide.' They tell their agent to 'List all collections.' The agent runs list_collections and shows the PM the structure, identifying potential overlaps or missing sections.
Support agent needs a quick answer across all products.
A support engineer gets a complex question about three different product lines. They tell their agent, 'Find all details on the return policy.' The agent runs search_content and aggregates answers from multiple, disparate product spaces.
Technical writer needs to check page hierarchy.
A technical writer wants to know if the 'Advanced Settings' section is nested under 'User Guide.' They ask the agent to 'List pages in the User Guide space.' The agent runs list_pages and provides the full, structured hierarchy, saving the writer from manual clicking.
The Tradeoffs
Calling tools sequentially
Trying to find a page by first calling list_organizations, then running list_spaces on the result, and then calling list_pages for every single space. This is slow, tedious, and requires constant manual handoffs.
→
Use the agent to orchestrate the calls. Ask the agent to 'Find all pages in the 'User Guide' space.' The agent handles the list_spaces and list_pages calls internally, giving you the final list directly.
Guessing the required scope
A user just wants to check one specific section, but they don't know if it's in a 'collection' or a 'space,' so they try calling list_collections and then list_spaces separately, wasting time.
→
Start by asking the agent to list all organizations (list_organizations). This gives you the top-level view. Then, narrow the scope by asking the agent to 'List spaces in the [Organization Name]'.
Ignoring user context
Assuming the agent can access all data without checking permissions, leading to failed calls when the account context is restricted.
→
Always ask the agent to run get_me first. This confirms the user profile and permission limits before the agent attempts to retrieve sensitive data using other tools.
When It Fits, When It Doesn't
Use this server if your core problem is knowledge retrieval or documentation structure mapping. You need to query vast, siloed knowledge bases (like GitBook) and get structured answers without clicking through UIs. Don't use this if your goal is simply to draft a single page, or if you only need to manage a single document's text. For drafting, use a pure LLM chat interface. If you need to verify who owns the documentation, use get_me first. If you need to know where the documentation is, use list_organizations and list_spaces to map the scope. If you need the actual text, use get_page or search_content.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by GitBook. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 8 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Sifting through documentation manually is a massive time sink.
Today, finding a specific piece of technical info means jumping between tabs: checking the main index, drilling down into the right space, navigating the page hierarchy, and then scrolling past dozens of related articles just to find the one paragraph you need. You spend more time navigating than reading.
With this MCP server, you tell your agent exactly what you need. It handles the whole process—it maps the hierarchy using `list_spaces` and `list_pages`, and it pulls the exact answer using `search_content`. You get the answer, period.
GitBook MCP Server: Structured access to your knowledge.
No more guessing which tool to use. The server manages the complexity of listing organizations (`list_organizations`), finding spaces (`list_spaces`), and finally listing pages (`list_pages`). The agent does the heavy lifting, chaining the calls together so you just get the result.
This gives you a full, programmatic view of your knowledge base. It's not just a search box; it's a structured API that lets you audit and navigate the entire system.
Common Questions About GitBook MCP
How do I use the search_content tool in GitBook MCP Server? +
You simply ask your agent to search for the content. You don't call the tool directly. Just tell the agent, 'Search my GitBook for X.' The agent handles the search_content call and provides matches and context.
Can I list all spaces in GitBook MCP Server? +
Yes. Your agent can run list_spaces after you specify the organization name. This gives you a list of all available documentation environments for that organization.
What is the difference between get_page and search_content? +
get_page reads the full content of one specific page. search_content finds relevant snippets across many pages, making it better for broad, quick research.
Do I need to use get_me before running any GitBook MCP Server tools? +
It's best practice. Running get_me first confirms the authenticated user's identity and permission levels, ensuring the agent won't fail midway through a complex operation due to access restrictions.
How do I use list_organizations to see all available knowledge sources? +
It lists every organization connected to your GitBook account. This helps you confirm all knowledge domains—like 'Product Alpha' or 'Support Docs'—are available for your agent to search.
What is the difference between list_spaces and list_pages? +
list_spaces gives you the main documentation silos (the 'containers'). list_pages dives inside a specific space to list all the individual documents or articles contained within it.
How does the get_page tool handle long content retrieval? +
The get_page tool pulls the full text of a single document. It handles large pages by returning the entire content block, letting your agent process it chunk by chunk.
Can I use get_me to verify my user permissions for document access? +
Yes, the get_me tool fetches your authenticated profile metadata. This gives your agent the name and email, which you can then use to check account context and permission boundaries.
Can my agent search for technical content across all my GitBook spaces? +
Yes. Use the 'search_content' tool. Provide the 'space_id' and your query. The agent will execute cross-page search operations within that GitBook namespace, returning matching snippets and relevant content natively.
How do I retrieve the full content of a specific documentation page via chat? +
Use the 'get_page' tool. You will need the 'space_id' and the 'page_id'. Your agent will read the full document content from GitBook, allowing you to summarize, analyze, or verify the information flawlessly.
Can I see how my product documentations are grouped through the agent? +
Absolutely. Use the 'list_collections' tool with your Organization ID. Your agent will retrieve the collections that group multiple spaces, helping you understand your documentation hierarchy natively.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Goldsky (Web3 Data Indexing & Subgraphs)
Manage Web3 data pipelines and subgraphs. List, create, and monitor real-time indexing pipelines directly from your AI agent.
GeTui / 个推
Massive notification push and LBS platform in China — manage campaigns, devices, and reports via AI.
ROC AUC Evaluator
Compute the exact Area Under the ROC Curve for binary classification predictions. Local, mathematically perfect, zero LLM estimation.
You might also like
IBGE Nomes
Access official Brazilian name frequency and ranking data from IBGE — explore popularity by decade, gender, and locality.
Nutritionix
Analyze food nutrition from natural language using the industry-leading NLP engine — type any meal description and get instant, precise calorie and macro data.
Replicate Extended
Automate machine learning workflows via Replicate — run models, manage predictions, and search for AI assets directly from any AI agent.