wikijs-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PORT | No | Port to listen on when TRANSPORT=http. | 3000 |
| LOG_LEVEL | No | pino log level: fatal, error, warn, info, debug, trace. | info |
| TRANSPORT | No | stdio or http. | stdio |
| WIKIJS_URL | Yes | Base URL of your Wiki.js instance (required). | |
| WIKIJS_TOKEN | Yes | API key from Administration → API Access (required). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_pagesA | Full-text search Wiki.js pages. Returns matching pages with id, title, path and locale. Use get_page to fetch content. |
| list_pagesC | List Wiki.js pages with optional ordering, tag and locale filters. |
| get_pageA | Fetch one page by numeric id OR by path (+ locale, default "en"). Optionally slice its content by line range with startLine/endLine: 0-based, inclusive on both ends. Default (0..-1) returns the whole page. Negative values count from the end, Python-style (-1 = last line, -2 = second-to-last). |
| get_page_treeA | Browse the page/folder hierarchy one level at a time. Omit parent for the root level; pass a folder item id as parent to descend. |
| list_tagsA | List all page tags, or search tags matching a query string. |
| create_pageA | Create a new Wiki.js page. Path must not start with the locale (use "guides/install", not "en/guides/install"). |
| update_pageA | Update an existing page by id. Only the provided fields change. Get the id from get_page, list_pages or search_pages first. |
| move_pageB | Move a page to a new path (and optionally another locale). Requires the manage:pages permission on the API key. |
| delete_pageA | Permanently delete a page by id. DESTRUCTIVE and irreversible — confirm with the user before calling this. |
| get_page_historyA | List the version history trail of a page (newest first), paginated. |
| get_page_versionA | Fetch the content of a specific historical version of a page. Optionally slice its content by line range with startLine/endLine: 0-based, inclusive on both ends. Default (0..-1) returns the whole version content. Negative values count from the end, Python-style (-1 = last line, -2 = second-to-last). |
| restore_page_versionA | Restore a page to a previous version from its history. The current state is preserved as a new history entry. |
| list_assetsA | List uploaded assets (images/files) and subfolders in an asset folder. folderId 0 is the root folder. |
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 13 tools
Each tool targets a distinct resource and action: pages vs tags vs assets, and within pages there are clear separations between listing, searching, fetching, creating, updating, moving, deleting, and history operations. Potential confusions (list_pages vs search_pages, get_page vs get_page_version) are resolved by clear descriptions.
All tool names follow a consistent verb_noun pattern with lowercase snake_case. The verbs (list, search, get, create, update, move, delete, restore) are uniformly used, and nouns clearly indicate the target resource (pages, tags, assets, page_tree, page_history).
13 tools is a well-scoped number for a Wiki.js content management server. The count covers core page lifecycle, search/navigation, history/restore, and auxiliary tag/asset listing without being excessive or too sparse.
The page lifecycle is fully covered (create, read, update, delete, move, list, search, tree, history, restore). The only notable gap is asset management, where only listing is provided (no upload/download/delete), but this is likely outside the tool's primary content-focused purpose.