MediaWiki MCP. Query any wiki, track its history, or update content.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
MediaWiki MCP Server connects your AI client directly to any wiki instance—like Wikipedia or a corporate knowledge base. It lets you query content, track page history via `get_page_revisions`, list categories, and even execute changes using tools like `edit_page` through natural language prompts.
What your AI agents can do
Create account
Creates a new user account on the wiki, requiring tokens fetched via get_tokens.
Edit page
Edits and updates an existing wiki page's content, requiring a CSRF token from get_tokens.
Get page categories
Lists all the categories associated with a specified wiki page.
Run a full-text search across all pages in the wiki to locate specific information.
Retrieve detailed revisions, metadata, and internal links for any given page.
List all pages in a category or enumerate every single page existing on the wiki instance.
Get an ordered list of recent edits, showing who changed what and when.
Create new user accounts or directly edit the content of existing wiki pages (requires tokens).
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
MediaWiki MCP Server: 13 Tools for Wiki Data Access
These tools let your agent interact with the wiki API—from searching content and mapping structure to editing pages and auditing history.
019e5d34create account
Creates a new user account on the wiki, requiring tokens fetched via `get_tokens`.
019e5d34edit page
Edits and updates an existing wiki page's content, requiring a CSRF token from `get_tokens`.
019e5d34get page categories
Lists all the categories associated with a specified wiki page.
019e5d34get page info
Fetches basic, structured information about a specific wiki page.
019e5d34get page links
Lists all internal links found within the body of a specified wiki page.
019e5d34get page revisions
Retrieves historical content, metadata, or specific revisions for any given wiki page.
019e5d34get site info
Gathers general settings and high-level information about the entire wiki site structure.
019e5d34get tokens
Fetches necessary tokens (like CSRF or account tokens) required for data modification actions.
019e5d34get user info
Retrieves details about the user who is currently authenticated to the wiki site.
019e5d34list all pages
Enumerates and lists every single page that exists within the entire wiki namespace.
019e5d34list category members
Retrieves a list of all pages belonging to a specified category name.
019e5d34list recent changes
Returns an ordered log of the most recent edits and changes made across the wiki site.
019e5d34search pages
Performs a full-text search query against the content of all pages in the wiki.
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 MediaWiki, 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
MediaWiki MCP Server connects your AI client straight into any wiki instance—whether it's Wikipedia or some corporate knowledge base you gotta deal with. Your agent doesn't just look at pages; it gets the whole scoop, letting you query content, map out site structure, and even make changes using natural language prompts.
Finding Stuff & Mapping Out the Site Structure
Your agent runs a full-text search across every page using search_pages to pinpoint exactly what info you need. Need to know if a specific piece of content exists at all? You can use get_page_info for basic details on any given wiki page, or run list_all_pages to enumerate every single namespace that's built into the site.
To figure out where things fit together, you get the full picture. Your agent pulls a list of all categories associated with a specific page using get_page_categories, and if you want to see everything in one spot, it lists all members within a category name via list_category_members. You also find every internal link baked into a page's body by calling get_page_links.
Digging Into History & Auditing Changes
If you need to know who changed what and when, your agent pulls an ordered log of recent edits using list_recent_changes, showing the details for every modification across the site. For deep dives, it retrieves historical content, metadata, or specific revisions for any page via get_page_revisions. You can also check out general settings and high-level info about the whole wiki setup with get_site_info.
Curious about who's logged in? Your agent fetches details about the user currently authenticated to the site using get_user_info. For a complete record, it lets you get basic information about the entire site structure itself.
Making Changes and User Management
Need write access? It ain't automatic. First, your agent has to grab necessary tokens—like CSRF or account credentials—by calling get_tokens. Once you've got those tokens, you can create a brand new user account on the wiki using create_account. If you gotta update content, your agent edits and updates an existing page's content directly via edit_page.
This server gives your AI client the keys to the kingdom. You get granular control over reading data—searching pages with search_pages, listing all pages with list_all_pages, mapping categories with get_page_categories, and checking out history with get_page_revisions—and you also gain the power to modify content by calling edit_page. You'll see your agent pull user data through get_user_info, grab tokens using get_tokens, or list recent changes across the board with list_recent_changes.
How MediaWiki MCP Works
- 1 First, subscribe to the server and enter your target MediaWiki API URL. If you need write access, fetch necessary credentials using
get_tokens. - 2 Next, instruct your AI agent on the specific data you need—for example, 'List all pages in the 'Finance' category.'
- 3 The agent executes the appropriate tool (e.g.,
list_category_members), and you receive a structured list of results.
The bottom line is: your AI client runs the specific MediaWiki API calls needed to pull or push data, bypassing manual web navigation entirely.
Who Is MediaWiki MCP For?
Documentation Specialists and Technical Writers who spend too much time manually verifying content consistency. Also targeting Backend Developers needing accurate project context from internal wikis, and Academic Researchers who need to compare data across multiple knowledge bases.
Uses get_page_revisions and list_recent_changes to audit documentation against the latest source of truth before publishing.
Runs list_category_members or search_pages to identify gaps in internal wikis, ensuring all required topics are documented and linked.
Checks site configurations with get_site_info or verifies API documentation by running specific queries against the wiki structure.
What Changes When You Connect
- Stop manual searching. Instead of clicking through pages to find a keyword, run
search_pagesand get immediate results across the entire knowledge base. - Audit documentation changes instantly. Use
list_recent_changesto see who edited what page, when, and why, without navigating the admin dashboard. - Understand content context deeply.
get_page_revisionslets you look at not just the current text, but every previous version of a page—critical for compliance checks. - Map the whole site's architecture. Running
list_all_pagesorlist_category_membersgives your agent an instant inventory of available documentation, no matter how big the wiki is. - Maintain data integrity. If you need to update a policy, use
get_tokensfirst, then runedit_page. This ensures you're authenticated before making changes.
Real-World Use Cases
Developer needs API context for a new feature.
A developer is building a microservice and needs to know the exact parameters used in an old design spec. They ask their agent: 'What are the core links on the Project X page, and what is its current status?' The agent calls get_page_links and then get_page_info, providing all necessary metadata for the developer to proceed without interrupting a subject matter expert.
Compliance officer needs to verify policy adherence.
The compliance team suspects that department policies are outdated. They ask: 'Show me all pages in the 'HR Policy' category and list the last five edits on any of them.' The agent calls list_category_members followed by list_recent_changes, giving the officer a full audit trail for immediate review.
Academic researcher needs to compare source material across multiple departments.
A student is writing about 'Climate Change' and needs input from three separate corporate wikis. Instead of logging into three different systems, they prompt their agent: 'Search all relevant pages for 'GHG Emissions'.' The agent runs search_pages against the connected wiki instance, consolidating results instantly.
Content team needs to update a master document.
The content lead must update the 'Product Roadmap' page. They first call get_tokens for write access. Then, they use edit_page and provide all new text sections in one go, ensuring the change is logged correctly.
The Tradeoffs
Treating the wiki like a search engine.
Asking the agent: 'Tell me everything about AI.' The agent might just return general links without structure.
→
Be specific. Instead of a broad query, use search_pages with keywords AND specify the namespace if possible. If you need structured data, ask for get_page_info on a known page.
Assuming write access without credentials.
The agent attempts to run edit_page and fails with an authentication error because it never ran the required setup steps.
→
Always start by calling get_tokens. This fetches necessary, time-sensitive tokens (like CSRF) that must be passed to any modification tool like edit_page.
Trying to get a full site map via manual clicks.
A user spends an hour clicking through the wiki's sidebar and category tabs just to list all available topics.
→
Run list_all_pages or, if you know the grouping, run list_category_members. This provides a complete enumeration instantly.
When It Fits, When It Doesn't
Use this server if your workflow relies on structured knowledge—meaning content needs history tracking (get_page_revisions), organizational structure mapping (categories/links), or modification rights. Don't use it if you just need general web browsing; that requires a search engine API. If you only want to read the latest articles, search_pages is sufficient. But if you need to know who wrote those articles and when, you must call list_recent_changes. Never forget: modification tools like edit_page require tokens first; they are not standalone calls.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by MediaWiki. 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 13 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Finding a specific piece of documentation shouldn't take five clicks.
Right now, finding the definitive source for a policy means opening the wiki, navigating by category sidebar, clicking through multiple index pages, and then reading dozens of search results to find the single correct document. It’s manual, slow, and you might miss an important revision.
With this MCP server, your agent runs `search_pages` instantly. You give it a topic—say, 'Q3 marketing compliance'—and it returns the exact page reference and context without any clicks or guesswork.
MediaWiki MCP Server: Structured data from `get_page_revisions`.
Manual auditing means comparing screenshots of different pages over time, making it nearly impossible to prove what the official text was on a specific date. You are left guessing based on memory or vague timestamps.
Now, calling `get_page_revisions` returns structured JSON data showing content snapshots and metadata for every version. It’s proof-grade history you can write code against.
Common Questions About MediaWiki MCP
How do I search the wiki using the `search_pages` tool? +
You simply tell your agent to run search_pages and provide the keywords. It executes a full-text query against all indexed content, returning relevant pages directly.
Do I need `get_tokens` before using `edit_page`? +
Yes. You must call get_tokens first to retrieve the necessary CSRF or account tokens. These temporary credentials are required by the edit_page tool for successful data modification.
`list_recent_changes` vs `get_page_revisions`, what's the difference? +
list_recent_changes gives you an ordered log of all recent edits across the whole site. get_page_revisions lets you deep-dive into the specific version history of one single page.
How can I find all related documentation for a topic? +
Use list_category_members. If you know the category name (e.g., 'API Documentation'), this tool will list every page that belongs to it, giving you a complete scope view.
What happens if I try to use the `edit_page` tool, but my account lacks proper permissions? +
The agent will return a specific API error code indicating insufficient write access. You need to verify your user role on the wiki or ensure the credentials used by your AI client have editor-level rights.
Is it efficient to use `list_all_pages` when I only want information about a specific subject? +
No, calling list_all_pages generates a massive list of slugs and is slow. For targeted data, always start by using search_pages or checking the contents of a known category with list_category_members.
What specific types of data can I retrieve about the current user using `get_user_info`? +
It provides core account details, including the username, join date, and sometimes last login information. This is useful for auditing or confirming who originally created a piece of content.
How does the `get_page_links` tool help me trace cross-references on a page? +
The tool lists every internal link found within the source code of that single page. It doesn't just provide titles; it shows exactly which other pages or sections are linked to from this location.
How do I get the actual text content of a specific wiki page? +
Use the get_page_revisions tool with the page title. You can specify rvprop as 'content' to retrieve the full text of the latest revision.
Can I see all pages that belong to a specific category like 'Physics'? +
Yes! Use the list_category_members tool and provide the category title (e.g., 'Category:Physics'). It will return a list of all pages filed under that category.
How do I check the general settings and namespaces of the wiki? +
Simply run the get_site_info tool. It fetches general information about the MediaWiki site, including its configuration and available namespaces.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Liveblocks
Manage real-time collaboration infrastructure—create rooms, authorize users, and inspect collaborative storage or threads directly from your AI agent.
Geekbot
Run asynchronous standups and team check-ins through Slack or Microsoft Teams that respect everyone timezone and schedule.
Slack Webhook Notifier
This MCP does exactly one thing: it sends messages to your Slack channels. That's its only function, and nothing else. Zero bloat, incredible for giving your AI agents a voice.
You might also like
Cincopa
Manage video and media galleries via Cincopa — track assets, trigger uploads, and monitor engagement directly from any AI agent.
United Airlines
Track United Airlines flights, schedules, routes, delays, and fleet data in real-time via AI agents.
Slack
Manage channels, users, and send messages on Slack with AI agents.