Atlassian Confluence MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PAT | No | API token (Cloud) or Personal Access Token (Server/DC) | |
| CONFLUENCE_BASE_URL | Yes | Root URL of your Confluence instance (required) | |
| CONFLUENCE_PASSWORD | No | Password or API token for Basic Auth | |
| CONFLUENCE_USERNAME | No | Your Atlassian account email (Cloud) or username (Server/DC basic auth) |
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 | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| confluence_test_connectionA | Test connection to Confluence API |
| confluence_get_spacesC | List all spaces in Confluence |
| confluence_get_spaceB | Get details of a specific space by key |
| confluence_get_space_by_keyB | Get details of a specific space by key |
| confluence_get_pagesA | List pages, optionally filtered by space |
| confluence_get_pageA | Get a specific page by ID (for very large pages, prefer confluence_get_page_body_chunk) |
| confluence_get_page_by_titleA | Get a page by its title within a space |
| confluence_get_page_body_chunkA | Read a page body in chunks by character offset/length (useful for very large pages) |
| confluence_searchA | Search for content using CQL (Confluence Query Language) |
| confluence_search_pagesB | Search for pages by text content |
| confluence_get_page_labelsA | Get labels attached to a page |
| confluence_get_page_commentsB | Get comments on a page |
| confluence_get_page_attachmentsB | Get attachments on a page |
| confluence_get_comment_attachmentsB | Get attachments on a comment |
| confluence_get_child_pagesA | Get child pages of a parent page |
| confluence_get_page_ancestorsB | Get ancestor pages of a page |
| confluence_get_current_userA | Get the current authenticated user |
| confluence_get_page_versionsC | Get version history of a page |
| confluence_get_page_versionA | Get a specific version of a page |
| confluence_get_page_propertiesA | Get content properties of a page |
| confluence_get_page_by_urlA | Get a page by its Confluence URL (e.g., https://confluence.example.com/display/SPACE/Page+Title or ?pageId=12345) |
| confluence_get_personal_space_keyA | Get the personal space key for the current authenticated user |
| confluence_get_space_homepageA | Get the homepage of a space |
| confluence_get_page_watchersB | Get users watching a page |
| confluence_get_page_permissionsB | Get page restrictions/permissions |
| confluence_get_recently_modifiedB | Get recently modified pages |
| confluence_get_page_tasksB | Get inline tasks from a page |
| confluence_export_pageA | Get export URL for a page (PDF or Word) |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Confluence Spaces | List of all Confluence spaces |
TDQS
Scored across 28 tools
Most tools target distinct resources, but there is clear overlap between confluence_get_space and confluence_get_space_by_key, which have identical descriptions. confluence_search and confluence_search_pages also blur together since both search for pages by different mechanisms.
Tool names consistently follow a confluence_verb_noun pattern and use snake_case throughout. Minor deviations like confluence_get_recently_modified lacking a noun and confluence_test_connection not using get are not seriously confusing.
28 tools is above the 25+ threshold and feels heavy for a set that is almost entirely read-only. Many niche endpoints like watchers, tasks, personal space key, and individual versions could be consolidated.
The tool surface is severely incomplete for a Confluence integration because it contains no create, update, delete, or write operations at all. Agents can read and search content but cannot perform any meaningful action on the Confluence instance.