mcp-mediawiki-crunchtools
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MEDIAWIKI_URL | Yes | Wiki base URL (e.g., https://en.wikipedia.org/w/) | |
| MEDIAWIKI_PASSWORD | No | Bot/user password | |
| MEDIAWIKI_USERNAME | No | Bot/user account for write operations | |
| MEDIAWIKI_HTTP_PASS | No | HTTP Basic Auth password | |
| MEDIAWIKI_HTTP_USER | No | HTTP Basic Auth username (.htaccess) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Secure MCP server for MediaWiki wikis. Search, read, create, edit, and manage wiki pages, categories, files, and more. Works with any MediaWiki instance (public or private).
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": true
} |
| resources | {
"subscribe": false,
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_toolC | Full-text search across the wiki. |
| get_page_toolB | Get page wikitext content by title. |
| get_page_html_toolC | Parse a wiki page to HTML. |
| list_pages_toolB | List wiki pages with optional prefix filter. |
| create_page_toolC | Create a new wiki page. Requires authentication. |
| edit_page_toolC | Edit an existing wiki page. Requires authentication. |
| delete_page_toolB | Delete a wiki page. Requires admin authentication. |
| move_page_toolC | Move (rename) a wiki page. Requires authentication. |
| list_categories_toolC | List all categories on the wiki. |
| get_category_members_toolC | Get pages in a category. |
| get_page_categories_toolB | Get categories that a page belongs to. |
| list_recent_changes_toolC | List recent edits on the wiki. |
| parse_wikitext_toolC | Parse raw wikitext to HTML. |
| get_site_info_toolA | Get wiki configuration and version info. |
| list_namespaces_toolA | List all namespaces on the wiki. |
| get_user_info_toolC | Get user details from the wiki. |
| list_user_contributions_toolC | List edits by a user. |
| get_file_info_toolB | Get file/image metadata from the wiki. |
| list_files_toolC | List files on the wiki. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 19 tools
Each tool targets a distinct resource+action pair (pages, categories, files, users, search, namespaces), so misselection is unlikely. The only fuzzy boundary is between get_page_html_tool (renders a page by title) and parse_wikitext_tool (renders raw wikitext), which sound similar until you read the input difference. get_page_tool vs get_page_html_tool is also a subtle distinction but handled by descriptions.
Every tool follows the same verb_noun_tool pattern (list_, get_, create_, edit_, delete_, move_, parse_) in consistent snake_case with a uniform _tool suffix. Pluralization for list operations (list_pages, list_files, list_categories) is applied predictably. No mixed conventions.
19 tools is slightly heavy but each one maps to a meaningful MediaWiki operation across pages, categories, files, users, and search. There is very little redundancy — nothing overlaps enough to be pruned. It sits near the upper edge of comfortable scoping for a single wiki API.
Page lifecycle is fully covered (list/get/create/edit/delete/move), but revision history and diffs are entirely absent, which is a core wiki workflow. Files and categories are read-only — no upload/delete file, no add/remove category from a page — leaving agents to dead-end on those write paths. User tooling is also read-only.