Supercharge your AI with Wiki.js. Manage your entire documentation structure via conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Wiki.js MCP connects your AI agent directly to your internal knowledge base. You can manage documentation by creating new pages, updating existing content, deleting outdated docs, and searching across every page instantly—all without opening a browser or writing API calls.
What your AI can do
Create page
Generates and publishes a brand new documentation page within your wiki instance.
Delete page
Removes an entire document from the knowledge base, marking it as deleted.
Get page
Retrieves all content and metadata for a specific page using its unique path.
List every single page available in the wiki to understand the scope of your knowledge base.
Run precise searches across the entire documentation library using keywords or phrases.
Retrieve the complete text and metadata for a page when you know its exact URL structure.
Generate and publish entirely new documentation articles with full formatting and metadata.
Modify the text or details of a page using its unique identifier without losing historical context.
Delete documentation that is no longer accurate, keeping your knowledge base clean and current.
Ask an AI about this
Compatible AI Apps
OAuth 2.0 Compatible Gemini Waiting for input…
Wiki.js: 6 Tools for Knowledge Base Management
These tools give your agent granular control over every aspect of your wiki content, allowing you to manage documents at the API level.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Wiki.js on VinkiusCreate Page
Generates and publishes a brand new documentation page within your wiki instance.
Delete Page
Removes an entire document from the knowledge base, marking it as deleted.
Get Page
Retrieves all content and metadata for a specific page using its unique path.
List Pages
Provides a full inventory of every available page title in the wiki.
Search Pages
Searches the content and metadata across all pages for specified keywords or phrases.
Update Page
Modifies the text, title, or status of an existing page using its ID.
Connect to your AI in seconds. Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token from
cloud.vinkius.com. For OAuth-protected servers, expand Advanced
settings to add credentials.
Start a conversation
Open a new chat. The Wiki.js integration is available immediately — no restart needed.
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 Wiki.js, then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ 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
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Wiki.js. 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 connection provides 6 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Finding specific information in your company wiki is an exercise in frustration.
Today, finding a single piece of correct information means clicking through several internal links. You navigate to the 'API' section, then drill down into 'V2 Guides,' and finally type keywords into a search bar that returns 47 results. You have to manually click each link, skim the first few paragraphs, and copy-paste the relevant snippet—a process that takes forever.
With this MCP, your agent bypasses all those clicks. Instead of navigating pages, you just ask your AI client for what you need. The agent immediately runs `search_pages` across the entire knowledge base, giving you a list of accurate results and letting you pinpoint the exact page path instantly.
The Wiki.js MCP gives you full control over content lifecycle management.
You manually create pages by using a markdown editor, then remember to set the metadata and publish it correctly, which often gets skipped. You also have to go into the system settings just to check if a page exists or needs fixing.
Now, you use the `create_page` tool to build the initial structure and `get_page` to verify its existence instantly. The agent manages the lifecycle from draft to published with single commands, guaranteeing your documentation stays accurate.
What your AI can actually do with this
Need to structure or update technical guides? This connector gives your AI agent full control over your Wiki.js instance. Instead of navigating multiple tabs or exporting data just to hand it off to someone else, your agent talks directly to the knowledge base. It can list every page you have, search for a specific concept across all content, and even create brand new sections with metadata attached.
You'll stop treating documentation like static files and start treating it like live, editable data.
If you run complex operations involving multiple internal systems—say, pulling customer records from one source and updating the corresponding guide in another—you can chain these tools together using Vinkius. Since your credentials pass through a zero-trust proxy, your keys never sit on disk, keeping your sensitive documentation safe while automating massive workflows.
019e390a-fb1c-7250-a0a7-2d4f01601d70 
Here's how it actually works
The bottom line is, you tell your agent what needs doing, and it executes the necessary steps directly against your wiki.
Subscribe to this MCP and input your Wiki.js API URL and key.
Your agent connects using the secure Vinkius platform, which handles all authentication in transit.
You prompt your AI client with a natural language request (e.g., 'Find documentation on X' or 'Create a page about Y').
Who is this actually for?
Content Managers who waste hours cross-referencing old guides. Technical Writers frustrated by version control issues. Support Engineers needing immediate access to complex internal procedures.
Needs to draft a new guide, search for existing boilerplate content using search_pages, and then publish it using create_page.
Must prune the wiki by listing all pages with list_pages, identifying stale content, and running delete_page on expired guides.
Needs to quickly find the most up-to-date procedure for a rare customer issue using get_page based on an article path.
What Changes When You Connect
Stop manually searching the GUI. If you need to find a guide, simply ask your agent and let it run search_pages across millions of words instantly.
Never lose content or struggle with versioning again. You can use get_page by its exact path to guarantee you are working on the correct source material every time.
Build automated documentation workflows. Combine this MCP with a ticketing system MCP—when an issue is logged, your agent can automatically pull details and draft a new page using create_page.
Keep your knowledge base clean. Instead of manually reviewing content for staleness, use list_pages to audit the structure and then run delete_page on guides that haven't been touched in years.
Save time writing repetitive docs. Use update_page to modify a single section (like legal disclaimers) across multiple related pages simultaneously.
See it in action
Onboarding new engineers
A new engineer needs the setup guide, but it's spread across three different documents. The agent runs list_pages, finds all 'onboarding' content, and compiles a single summary of key steps by pulling data from multiple sources using get_page.
Updating compliance requirements
Compliance mandates change the privacy policy. Instead of manually finding every mention, the agent runs a targeted search with search_pages, identifies all relevant guides, and then uses update_page to apply the new text across the board.
Auditing content accuracy
The team suspects old documentation is misleading. The agent runs list_pages to get a full inventory, flags pages with low traffic metadata, and helps prune them by running delete_page on the oldest guides.
Creating API reference docs
A developer finished a new microservice. They tell their agent to generate an initial guide. The agent uses create_page to build the skeleton, then uses get_page and update_page iteratively until the documentation is complete.
The honest tradeoffs
Trying to delete content manually
A user logs into Wiki.js, finds a page they want gone, clicks the trashcan icon, and then forgets about it because it's buried in archives.
Use the delete_page tool through your agent. This gives you an immediate audit trail of the removal action, proving exactly when and why the content was purged.
Searching only by title
A user searches for 'Authentication' but the guide is titled 'Security Protocols'. The manual search fails because it misses the keyword inside the document body.
Always use search_pages. This tool looks at the actual content, not just the page titles or metadata, guaranteeing you find what you need.
Updating docs via copy/paste
A writer copies a block of text and pastes it into an old guide. The formatting breaks, and they have no record of when that content was last reviewed or by whom.
Use update_page. This tool handles the revision process directly via API, maintaining clean version history while incorporating your new text.
When It Fits, When It Doesn't
Use this MCP if managing structured, interconnected documentation is the core task. You need to read specific content (get_page), find general topics (search_pages), or maintain a clear record of page existence and status (list_pages). Don't use it if your goal is simply having an unstructured chat about knowledge; in that case, you just talk to your agent without calling any tools. If your process involves complex data manipulation (e.g., combining customer IDs with API documentation), then chaining this MCP with a CRM or Database MCP through Vinkius will give you the full automation cycle.
Questions you might have
How do I find a page using the Wiki.js MCP? Should I use `get_page` or `search_pages`? +
Search_pages is for finding topics across many pages, while get_page requires you to know the exact URL path of one specific document.
Can I automate deleting old wiki content using delete_page? +
Yes. You can set up a workflow that uses your agent to list pages via list_pages, filter them for age, and then execute delete_page on the identified outdated documents.
What if I need to modify content but don't know the page ID? Should I use update_page? +
No. You should first run get_page or search_pages to find the correct path and metadata, which will provide the necessary identifiers for update_page.
Is there a tool to list all pages in the Wiki.js MCP? +
Yes, use list_pages. This function provides a complete inventory of every page title and path, letting you audit your entire knowledge base structure at a glance.
When I use `get_page`, how do I fetch content written in a specific language or locale? +
You must provide both the URL path and the required locale code when calling get_page. The tool uses this pairing to pull the correct, localized version of the raw content and metadata.
If I use `update_page`, how do I control whether a page is published or kept as a private draft? +
You specify the desired publishing status directly in the update payload. This lets you manage visibility, ensuring content is live, published, or restricted to internal drafts before calling update_page.
Does `create_page` allow me to set initial metadata like a page owner or category tags? +
Yes, you include a full metadata payload when running create_page. This ensures the new documentation is correctly attributed and categorized right from its initial creation.
What happens if I run `list_pages` but only have read-only access to the wiki? +
The tool respects your API key's defined scope. If you lack directory listing permissions, the call will fail or return an empty set of results, protecting unauthorized discovery.
Can I search for specific keywords across all my wiki pages? +
Yes! Use the search_pages tool with your query string. The agent will scan all content and return relevant page matches including titles and paths.
How do I fetch the content of a specific page if I have the URL path? +
Use the get_page tool. Simply provide the path (e.g., 'home' or 'engineering/setup') and the agent will retrieve the full content and metadata for that specific page.
Is it possible to modify an existing page's content? +
Absolutely. Use the update_page tool by providing the unique page ID and the new content or title you wish to apply. The agent will handle the update via the Wiki.js API.
We've already built the connector for Wiki.js. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 6 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.