mediawiki-mcp-server
The MediaWiki MCP server lets you search, read, edit, and analyze MediaWiki wikis via AI assistants or a CLI.
Search & Read
Full-text search, retrieve page content (wikitext or HTML), get sections/summaries, batch-fetch pages, search within a page, get metadata, parse wikitext to HTML, and fuzzy title resolution.
Link Analysis
Get/verify external links, find broken internal links, and get backlinks ("What links here").
Content Discovery & Quality
Find related pages, orphaned pages, similar/duplicate pages, stale pages, compare topic coverage across pages, check terminology against a glossary, find missing translations, and run a comprehensive wiki health audit.
History & Changes
View revision history, compare revisions with diffs, get recent wiki-wide changes (aggregated by user/page/type), and view user contributions.
Listing & Discovery
List pages, categories, category members, users by group, and get overall wiki statistics.
Write & Edit (require authentication)
Create/edit pages, find-and-replace (single page or bulk across pages/categories), apply formatting (bold, italic, strikethrough, code, underline), move/rename pages, manage categories, and upload files from URLs.
File Search
Search text within wiki-hosted PDFs, TXT, MD, CSV, JSON, XML, and HTML files.
Markdown Conversion
Convert Markdown to MediaWiki markup with optional themes (Tieto, neutral, dark), changelog reordering, and emoji prettification.
Safety & Output
All write operations support preview mode to verify changes before applying. Every operation supports JSON output for machine-readable or AI agent consumption.
Enables AI assistants to search, read, and edit content on Fandom-hosted wikis using natural language through MediaWiki API integration.
Provides specialized configuration for connecting to and interacting with Tieto's Public 360° Wiki for content retrieval and search.
Allows users to search Wikipedia, summarize pages, view revision histories, and check for broken links via the Wikipedia API.
MediaWiki MCP Server
Connect your AI assistant to any MediaWiki wiki, or script it directly from the terminal. Search, read, and edit wiki content using natural language or the wiki CLI.
Three ways to use it:
MCP server — Claude Desktop, Claude Code, Cursor, ChatGPT, n8n, VS Code, Google ADK, or any MCP-compatible tool. See SETUP.md.
wikiCLI — same API client, same auth, no AI needed. For shell pipelines, CI checks, cron jobs. See CLI.md.Claude Code plugin —
/plugin marketplace add olgasafonova/mediawiki-mcp-serveradds wiki skills directly to Claude Code. See .claude-plugin/README.md.
Documentation
Document | What it covers |
Get running in 2 minutes | |
Per-tool configuration (Claude Desktop, Cursor, ChatGPT, n8n, VS Code, Google ADK) | |
| |
Full tool reference (43 tools by category) | |
HTTP transport, security, endpoints, env vars | |
Connect to Tieto's Public 360° Wiki (beginner-friendly) | |
Detailed workflows by persona | |
System design | |
How to contribute | |
Security policies | |
Version history |
Related MCP server: WikiMCP
What Can You Do?
The same operation works from a prompt (via MCP) or directly in your shell (via the wiki CLI):
Goal | Prompt your AI | From your terminal |
Search the wiki | "What does our wiki say about onboarding?" |
|
Read a page | "Show me the Getting Started page" |
|
Find broken links | "Are there broken links on the Docs page?" |
|
Find stale content | "Which pages haven't been updated in 90 days?" |
|
Cross-link suggestions | "What pages are similar to the API Reference?" |
|
Audit wiki health | "Run a health check on the wiki" |
|
Publish markdown | "Publish this README to the wiki" |
|
Strike a name | "Strike out John Smith on the Team page" |
|
For the full tool list, see TOOLS.md. The CLI returns typed exit codes for CI-friendly branching — see CLI.md.
30-Second Start
Get the binary. Download from Releases or
go build -o mediawiki-mcp-server .(requires Go 1.24+).Find your wiki API URL — usually
https://your-wiki.com/api.php. Wikipedia ishttps://en.wikipedia.org/w/api.php. VisitSpecial:Versionto confirm.Wire it up to your AI tool — see SETUP.md for the configuration that matches your client.
Reading public wikis works without login. Private/corporate wikis and editing require a bot password — SETUP.md#editing-wiki-pages.
Example Prompts
📖 More examples: see WIKI_USE_CASES.md for detailed workflows by persona (content editors, documentation managers, developers).
Search and read:
"What does our wiki say about deployment?"
"Give me a quick overview of the Configuration page"
"Get the content of Main Page, FAQ, and Setup all at once"
Track changes:
"What pages were updated this week?"
"Show me the diff between the last two versions"
"Who are the most active editors this month?"
Check quality:
"Are there broken links on this page?"
"Find orphaned pages with no links to them"
"Find pages similar to the Installation Guide"
"Find pages not updated in the last 90 days"
Page management (requires auth):
"Rename 'Old Guide' to 'Updated Guide'"
"Strike out John Smith on the Team page"
"Replace 'version 2.0' with 'version 3.0' on Release Notes"
File uploads and search (requires auth):
"Upload this image from URL to the wiki"
"Upload this local file to the wiki" (the agent base64-encodes the bytes; no public URL needed)
"Search for 'budget' in File:Annual-Report.pdf"
Convert Markdown:
"Convert this README to wiki format"
"Convert with Tieto branding and CSS" (use theme="tieto", add_css=true)
PDF Search Setup
PDF search requires the pdftotext tool from poppler-utils. Text file search (TXT, MD, CSV, etc.) works without any dependencies.
Platform | Install Command |
macOS |
|
Ubuntu/Debian |
|
RHEL/CentOS |
|
Windows |
|
Windows alternative: Download binaries from poppler-windows releases and add to PATH.
Verify installation:
pdftotext -vCompatibility
Platform | Transport | Status |
Claude Desktop (Mac/Windows) | stdio | ✅ Supported |
Claude Code CLI | stdio | ✅ Supported |
Cursor | stdio | ✅ Supported |
VS Code | stdio | ✅ Supported |
ChatGPT | HTTP | ✅ Supported |
n8n | HTTP | ✅ Supported |
Google ADK | stdio / HTTP | ✅ Supported |
Works with any wiki: Wikipedia, Fandom, corporate wikis, or any MediaWiki installation.
Troubleshooting
Common issues and fixes live in SETUP.md#troubleshooting. For deeper diagnostics, the HTTP server exposes /health, /ready, and /status endpoints — see DEPLOYMENT.md.
Development
Build, test, and contribute: see CONTRIBUTING.md. For the system design, package layout, and data flow, see ARCHITECTURE.md.
Run the full check before pushing:
go test -race -failfast ./...
golangci-lint run ./...Integration tests run against a real MediaWiki instance via docker-compose.test.yml. See CONTRIBUTING.md for the workflow.
Like This Project?
If this server saved you time, consider giving it a ⭐ on GitHub. It helps others discover the project.
More MCP Servers
Check out my other MCP servers:
Server | Description | Stars |
Access GLEIF LEI database. Look up company identities, verify legal entities. | ||
Control Miro whiteboards with AI. Boards, diagrams, mindmaps, and more. | ||
Access Nordic business registries. Look up companies across Norway, Denmark, Finland, Sweden. | ||
Talk to your ProductPlan roadmaps. Query OKRs, ideas, launches. | ||
Nordic grocery deal hunting. Find offers, plan meals, track spending. | ||
Go library for SEP-2127 Server Cards. Pre-connect discovery for MCP servers. |
License
MIT License
Credits
Built with Go MCP SDK
Powered by MediaWiki API
Maintenance
Latest Blog Posts
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/olgasafonova/mediawiki-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server