# Confluence Team wiki. Create, read, and update pages and their children, attach files, comment, label, and search across a whole site. | Property | Value | | --- | --- | | Slug | `confluence` | | Definition version | `0.1.0` | | Base URL | `https://api.atlassian.com` | | Auth schemes | `oauth2` | | Action tools | 30 | | By class | 16 read, 12 write, 2 destructive | | Triggers | 0 | | Provider rate limit | not declared, so outbound calls are unpaced | ## Measured routing accuracy 42 golden cases replayed through the router over the whole index: measured over corpus `ea4f12ad2948` (65 toolkits, 2283 tools indexed and 13 declared uncallable), 30 cases written by hand and 12 cases from the paraphrase pass. A case counts as top-1 when its gold tool ranked first and top-8 when it reached the slate at all. | Measure | Cases | Share | | --- | --- | --- | | top-1 | 18/42 | 42.9% | | top-8 | 36/42 | 85.7% | The sweep is offline: the reranker is a deterministic identity fake that returns candidates in the order retrieval produced them, so top-1 measures retrieval order rather than a reranked slate. `just eval-live` measures the same cases through the live reranker. ## Authentication Connect an entity with `ConnectionsService.InitiateConnection`, naming this toolkit's slug. Credentials stay in the connections vault; callers hold connected-account ids only. ### `oauth2` | Property | Value | | --- | --- | | Authorization URL | `https://auth.atlassian.com/authorize` | | Token URL | `https://auth.atlassian.com/oauth/token` | | Default scopes | `read:confluence-content.all`, `write:confluence-content`, `read:confluence-space.summary`, `write:confluence-space`, `read:confluence-user`, `search:confluence`, `offline_access` | | Refresh tokens | yes, the refresh daemon renews ahead of expiry | ## Tools 30 action tools. The catalog-wide slug is `confluence.`, which is what `search_tools` returns and `call_tool` takes. ### `confluence.list_spaces` List or browse the spaces (wikis) on a Confluence site, optionally filtered by key or type. Use when the user asks what spaces exist or wants to find a space's id. For one already-known space use get_space. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/spaces`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `keys` | string | no | Comma-separated space keys to filter to, e.g. "ENG,MKT". Omit to list all. | | `limit` | integer | no | Page size, default 25. | | `status` | string, one of current, archived | no | Restrict to current or archived spaces. Defaults to current. | | `type` | string, one of global, personal | no | Restrict to global (team) or personal spaces. Omit for both. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].homepageId` | string | | | `results[].id` | string | | | `results[].key` | string | | | `results[].name` | string | | | `results[].status` | string | | | `results[].type` | string | | Also retrieved by: "what wikis do we have", "show me every team workspace", "which confluence areas exist", "browse the available wikis". ### `confluence.get_space` Get one Confluence space's (wiki's) detail by id: key, name, type, and its homepage id. Use when the user names a specific space. For the whole list use list_spaces. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/spaces/{{params.space_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `space_id` | string | **yes** | Confluence space id, from list_spaces. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `homepageId` | string | | | `id` | string | | | `key` | string | | | `name` | string | | | `status` | string | | | `type` | string | | Also retrieved by: "tell me about this workspace", "pull up the engineering wiki area", "what's this team's home base". ### `confluence.create_space` Create a new Confluence space (wiki) with a unique key and name. Use when the user wants to start, set up, or spin up a new team space, not a page inside one. For a page use create_page instead. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/api/v2/spaces`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `description` | object | no | Optional space description, e.g. {"value": "Engineering docs", "representation": "plain"}. Omit for no description. | | `description.representation` | string, one of plain, view | no | | | `description.value` | string | no | | | `key` | string | **yes** | Unique space key, e.g. "ENG". Uppercase letters and digits. | | `name` | string | **yes** | Space display name. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `id` | string | | | `key` | string | | | `name` | string | | | `status` | string | | Also retrieved by: "stand up a fresh wiki for the team", "I need somewhere new to keep docs", "open a workspace for this project", "set up a home for our notes". ### `confluence.update_space` Rename a Confluence space or edit its description. Use when the user wants to edit or update a space's own details, not the pages inside it. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `PUT /ex/confluence/{{params.cloud_id}}/wiki/api/v2/spaces/{{params.space_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `description` | object | no | New space description, e.g. {"value": "Engineering docs", "representation": "plain"}. Omit to leave unchanged. | | `description.representation` | string, one of plain, view | no | | | `description.value` | string | no | | | `name` | string | no | New space display name. Omit to leave unchanged. | | `space_id` | string | **yes** | Confluence space id to edit, from list_spaces. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `id` | string | | | `key` | string | | | `name` | string | | Also retrieved by: "rename our team's wiki area", "fix the blurb on our workspace", "change what this wiki is called". ### `confluence.create_page` Create a new Confluence page in a space, optionally under a parent page. Use when the user wants to write, add, or start a new wiki page or doc. Pass parent_id to make it a child page under an existing one; omit it for a top-level page in the space. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `body_value` | string | **yes** | Page content in Confluence storage format (XHTML-based markup), e.g. "

Hello world

". | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `parent_id` | string | no | Id of the page this becomes a child of. Omit for a top-level page in the space. | | `space_id` | string | **yes** | Confluence space id the page belongs to, from list_spaces. | | `status` | string, one of current, draft | no | Publish immediately (current) or save as a draft. Defaults to current. | | `title` | string | **yes** | Page title. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.webui` | string | | | `id` | string | | | `parentId` | string | | | `spaceId` | string | | | `status` | string | | | `title` | string | | | `version` | object | | | `version.number` | integer | | Also retrieved by: "write up a new doc", "jot down some notes on the wiki", "start a fresh article under this section", "draft something for the team to read". ### `confluence.get_page` Get one Confluence wiki page's detail and content by id: title, space, parent, current body, and version number. Use when the user asks to read, open, or pull up a specific Confluence page by its id. For a page found only by title use search_pages_by_title first. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `body_format` | string, one of storage, atlas_doc_format, view | no | Representation to return the body in. Defaults to storage. | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `page_id` | string | **yes** | Confluence page id. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.webui` | string | | | `authorId` | string | | | `body` | object | | | `body.storage` | object | | | `body.storage.value` | string | | | `createdAt` | string (date-time) | | | `id` | string | | | `parentId` | string | | | `spaceId` | string | | | `status` | string | | | `title` | string | | | `version` | object | | | `version.createdAt` | string (date-time) | | | `version.message` | string | | | `version.number` | integer | | Also retrieved by: "pull up that doc", "show me what's written on this article", "open the wiki entry for this", "read what's on that write-up", "what's on that confluence page". ### `confluence.update_page` Edit an existing Confluence wiki page's title or content, or move it under a different parent. Use when the user wants to change, edit, rewrite, or retitle a Confluence page. Requires the page's current version number, from get_page, incremented by one. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `PUT /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `body_value` | string | **yes** | New page content in Confluence storage format (XHTML-based markup), replacing the whole body. | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `page_id` | string | **yes** | Confluence page id to edit. | | `parent_id` | string | no | New parent page id, to move this page. Omit to leave it where it is. | | `status` | string, one of current, draft | no | Defaults to current. | | `title` | string | **yes** | New (or unchanged) page title. | | `version_number` | integer | **yes** | New version number. Must be exactly one greater than the page's current version.number, from get_page. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `id` | string | | | `title` | string | | | `version` | object | | | `version.number` | integer | | Also retrieved by: "fix the wording on this doc", "rewrite this article", "I need to correct what's written here", "refresh the content on this write-up", "edit that confluence page". ### `confluence.delete_page` Delete a Confluence page. Use when the user wants to delete, remove, or trash a page. By default it moves to the space's trash and can be restored there; pass purge true to permanently delete it right away. Irreversible once purged. Class `destructive` (held by the mutation gate until the call's own arguments carry `"confirm": true`). No scopes beyond the connection's defaults. Calls `DELETE /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `page_id` | string | **yes** | Confluence page id to delete. | | `purge` | boolean | no | Skip the trash and permanently delete right away. Default false. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `deleted` | boolean | | Also retrieved by: "get rid of this doc for good", "trash that write-up", "throw this article away", "I don't need this page anymore". ### `confluence.list_pages_in_space` List the pages in one Confluence space, optionally filtered by status. Use when the user asks what pages exist in a space or wants to browse its contents. For a page's child pages specifically, use list_child_pages. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/spaces/{{params.space_id}}/pages`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `space_id` | string | **yes** | Confluence space id, from list_spaces. | | `status` | string, one of current, archived, draft, trashed | no | Restrict to pages in this status. Defaults to current. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].authorId` | string | | | `results[].createdAt` | string (date-time) | | | `results[].id` | string | | | `results[].parentId` | string | | | `results[].status` | string | | | `results[].title` | string | | Also retrieved by: "what's written in this workspace", "show me everything under this wiki area", "browse the docs in this team space". ### `confluence.search_pages_by_title` Find Confluence pages across a site (or one space) by exact title. Use when the user names a page but you only have its title, not its id. For fuzzy keyword or full-text search use search_content instead. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `space_id` | string | no | Restrict the search to this space id. Omit to search the whole site. | | `title` | string | **yes** | Exact page title to match. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].id` | string | | | `results[].spaceId` | string | | | `results[].status` | string | | | `results[].title` | string | | Also retrieved by: "find the doc called onboarding guide", "I know the name but not where it lives", "look up an article by its exact heading". ### `confluence.list_child_pages` List the direct child pages nested under a Confluence page. Use when the user asks what sub-pages, nested pages, or children a page has. For every page in the whole space use list_pages_in_space. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/children`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `page_id` | string | **yes** | Confluence parent page id. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].id` | string | | | `results[].parentId` | string | | | `results[].status` | string | | | `results[].title` | string | | Also retrieved by: "what's nested under this doc", "show me the sub-sections of this article", "what write-ups branch off this one". ### `confluence.list_attachments` List the files attached to a Confluence page: images, PDFs, spreadsheets, and other uploads. Use when the user asks what's attached to a page or wants to find an attachment by filename. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/attachments`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `filename` | string | no | Filter to attachments matching this exact filename. | | `limit` | integer | no | Page size, default 25. | | `media_type` | string | no | Filter to attachments of this MIME type, e.g. "image/png". | | `page_id` | string | **yes** | Confluence page id. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].downloadLink` | string | | | `results[].fileSize` | integer | | | `results[].id` | string | | | `results[].mediaType` | string | | | `results[].title` | string | | | `results[].webuiLink` | string | | Also retrieved by: "what files are stuck to this doc", "show me the uploads on this article", "what's been dropped onto this write-up". ### `confluence.get_attachment` Get one Confluence attachment's detail by id: filename, media type, size, and download link. Use when the user names a specific attached file rather than browsing a page's whole attachment list. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/attachments/{{params.attachment_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `attachment_id` | string | **yes** | Confluence attachment id, from list_attachments. | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `downloadLink` | string | | | `fileSize` | integer | | | `id` | string | | | `mediaType` | string | | | `title` | string | | | `version` | object | | | `version.number` | integer | | | `webuiLink` | string | | Also retrieved by: "grab the details on that uploaded file", "where can I download this attached image", "info on the pdf someone put on the page". ### `confluence.upload_attachment` Attach a new file to a Confluence page. Use when the user wants to add, upload, or attach a file, image, or document to a page. Content is binary data, base64-encoded. To replace an existing attachment's content, use update_attachment_data instead. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/rest/api/content/{{params.page_id}}/child/attachment`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `comment` | string | no | Optional short comment describing the attachment. | | `content` | string | **yes** | File content, base64-encoded. | | `filename` | string | **yes** | File name including extension, e.g. "roadmap.pdf". | | `page_id` | string | **yes** | Confluence page id to attach the file to. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `results` | array of object | | | `results[].extensions` | object | | | `results[].extensions.fileSize` | integer | | | `results[].extensions.mediaType` | string | | | `results[].id` | string | | | `results[].title` | string | | Also retrieved by: "stick this file onto the doc", "drop an image into the article", "add this pdf to the write-up". ### `confluence.update_attachment_data` Replace an existing Confluence attachment's binary content, keeping its filename and id. Use when the user wants to overwrite, update, or upload a new version of a file already attached to a page. Content is binary data, base64-encoded. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/rest/api/content/{{params.page_id}}/child/attachment/{{params.attachment_id}}/data`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `attachment_id` | string | **yes** | Confluence attachment id to replace, from list_attachments. | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `comment` | string | no | Optional short comment describing the new version. | | `content` | string | **yes** | New file content, base64-encoded. | | `page_id` | string | **yes** | Confluence page id the attachment belongs to. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `extensions` | object | | | `extensions.fileSize` | integer | | | `id` | string | | | `title` | string | | Also retrieved by: "swap in a newer version of this file", "replace what's attached with the latest copy", "overwrite that uploaded document". ### `confluence.delete_attachment` Permanently delete a file attached to a Confluence page. Use when the user wants to delete, remove, or get rid of an attached file, not the whole page. Irreversible. Class `destructive` (held by the mutation gate until the call's own arguments carry `"confirm": true`). No scopes beyond the connection's defaults. Calls `DELETE /ex/confluence/{{params.cloud_id}}/wiki/api/v2/attachments/{{params.attachment_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `attachment_id` | string | **yes** | Confluence attachment id to delete, from list_attachments. | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `purge` | boolean | no | Skip the trash and permanently delete right away. Default false. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `deleted` | boolean | | Also retrieved by: "remove this file from the doc", "get rid of the attached image", "I want that upload gone for good". ### `confluence.list_footer_comments` List the footer comments on a Confluence page: the discussion thread at the bottom of the page, oldest first. Use when the user wants to read the general discussion on a page. For comments anchored to a specific highlighted bit of text, use list_inline_comments. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/footer-comments`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `page_id` | string | **yes** | Confluence page id. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].authorId` | string | | | `results[].body` | object | | | `results[].body.storage` | object | | | `results[].createdAt` | string (date-time) | | | `results[].id` | string | | | `results[].status` | string | | Also retrieved by: "what did people say at the bottom of this doc", "read the general discussion on this article", "show me the thread under this write-up". ### `confluence.create_footer_comment` Post a footer comment on a Confluence page, or reply to one already there. Use when the user wants to comment, reply, or leave general feedback on a page. For a comment anchored to specific text, use create_inline_comment instead. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/api/v2/footer-comments`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `body_value` | string | **yes** | Comment text in Confluence storage format (XHTML-based markup). | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `page_id` | string | **yes** | Confluence page id to comment on. | | `parent_comment_id` | string | no | Comment id to reply to. Omit to start a new thread. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `createdAt` | string (date-time) | | | `id` | string | | | `status` | string | | Also retrieved by: "leave a note at the bottom of this doc", "chime in on the discussion for this article", "reply to the thread on this write-up". ### `confluence.list_inline_comments` List the inline comments on a Confluence page: comments anchored to a specific highlighted selection of text, oldest first. Use when the user wants feedback tied to a particular passage. For the page's general discussion thread, use list_footer_comments. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/inline-comments`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `page_id` | string | **yes** | Confluence page id. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].authorId` | string | | | `results[].body` | object | | | `results[].body.storage` | object | | | `results[].createdAt` | string (date-time) | | | `results[].id` | string | | | `results[].status` | string | | Also retrieved by: "what feedback is anchored to that paragraph", "who highlighted and commented on this passage", "show me the annotations on this text". ### `confluence.create_inline_comment` Post a comment anchored to a specific highlighted selection of text on a Confluence page. Use when the user wants to comment on, flag, or react to one particular passage rather than the page as a whole. For a general comment use create_footer_comment instead. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/api/v2/inline-comments`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `body_value` | string | **yes** | Comment text in Confluence storage format (XHTML-based markup). | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `page_id` | string | **yes** | Confluence page id to comment on. | | `text_selection` | string | **yes** | The exact page text this comment is anchored to. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `createdAt` | string (date-time) | | | `id` | string | | | `status` | string | | Also retrieved by: "flag this specific sentence with a note", "react to just this highlighted bit", "annotate that particular passage". ### `confluence.get_comment` Get one Confluence comment's detail by id, footer or inline: its author, text, and version. Use when the user asks about one specific comment rather than a page's whole thread. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/comments/{{params.comment_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `comment_id` | string | **yes** | Confluence comment id, from list_footer_comments or list_inline_comments. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `authorId` | string | | | `body` | object | | | `body.storage` | object | | | `body.storage.value` | string | | | `createdAt` | string (date-time) | | | `id` | string | | | `pageId` | string | | | `status` | string | | | `version` | object | | | `version.number` | integer | | Also retrieved by: "pull up what exactly someone wrote there", "show me the details on that one remark", "what did that single reply actually say". ### `confluence.update_comment` Edit the text of a Confluence comment already posted, footer or inline. Use when the user wants to fix, correct, or update something they or someone else wrote in a comment. Requires the comment's current version number, from get_comment, incremented by one. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `PUT /ex/confluence/{{params.cloud_id}}/wiki/api/v2/comments/{{params.comment_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `body_value` | string | **yes** | Replacement comment text in Confluence storage format. | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `comment_id` | string | **yes** | Confluence comment id to edit. | | `version_number` | integer | **yes** | New version number. Must be exactly one greater than the comment's current version.number, from get_comment. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `id` | string | | | `version` | object | | | `version.number` | integer | | Also retrieved by: "fix a typo in what I wrote earlier", "correct my earlier remark", "edit that note I left". ### `confluence.delete_comment` Remove a comment from a Confluence page, footer or inline. Use when the user wants to delete or remove a comment, not the whole page. For deleting the page itself use delete_page. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `DELETE /ex/confluence/{{params.cloud_id}}/wiki/api/v2/comments/{{params.comment_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `comment_id` | string | **yes** | Confluence comment id to delete. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `deleted` | boolean | | Also retrieved by: "take down that remark", "erase what I said there", "pull that note off the doc". ### `confluence.list_page_labels` List the labels (tags) applied to a Confluence page. Use when the user asks what a page is tagged with. To add or remove a label use add_label or remove_label. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/labels`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `page_id` | string | **yes** | Confluence page id. | | `prefix` | string, one of global, my, team | no | Restrict to labels of this namespace. Omit for all. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].id` | string | | | `results[].name` | string | | | `results[].prefix` | string | | Also retrieved by: "what tags are on this doc", "how is this article categorized", "show me the topic tags for this write-up". ### `confluence.add_label` Tag a Confluence page with a label. Use when the user wants to tag, label, or categorize a page. To see what's already applied use list_page_labels; to remove one use remove_label. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `POST /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/labels`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `name` | string | **yes** | Label text, e.g. "roadmap". | | `page_id` | string | **yes** | Confluence page id to tag. | | `prefix` | string, one of global, my, team | no | Label namespace. Defaults to global. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `id` | string | | | `name` | string | | | `prefix` | string | | Also retrieved by: "tag this doc as a roadmap item", "mark this article under a category", "slap a label on this write-up". ### `confluence.remove_label` Remove a label from a Confluence page. Use when the user wants to untag, unlabel, or remove a tag from a page, not delete the page itself. Class `write` (writes, no confirmation needed). No scopes beyond the connection's defaults. Calls `DELETE /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/labels/{{params.label_id}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `label_id` | string | **yes** | Label id to remove, from list_page_labels. | | `page_id` | string | **yes** | Confluence page id the label is on. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `removed` | boolean | | Also retrieved by: "untag this doc", "take that category off the article", "strip the tag from this write-up". ### `confluence.search_content` Search across an entire Confluence site using CQL (Confluence Query Language), e.g. 'space = "ENG" AND type = page AND text ~ "roadmap"'. Use for any fuzzy, keyword, or cross-space search. For an exact known title use search_pages_by_title instead. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/rest/api/search`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cql` | string | **yes** | CQL query, e.g. 'space = "ENG" AND type = page AND text ~ "roadmap" ORDER BY lastmodified DESC'. | | `limit` | integer | no | Page size, default 25. | | `start` | integer | no | Offset into the result list, default 0. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `limit` | integer | | | `results` | array of object | | | `results[].content` | object | | | `results[].content.id` | string | | | `results[].content.space` | object | | | `results[].content.title` | string | | | `results[].content.type` | string | | | `results[].excerpt` | string | | | `results[].lastModified` | string (date-time) | | | `results[].url` | string | | | `start` | integer | | | `totalSize` | integer | | Also retrieved by: "hunt across the whole wiki for mentions of pricing", "find anything anywhere that talks about the launch", "dig through every space for this keyword". ### `confluence.list_page_versions` List a Confluence page's version history, newest first: who changed it, when, and their edit message. Use when the user asks about a page's history, past edits, or revisions. For one specific past version's content use get_page_version. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/versions`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `cursor` | string | no | Cursor from a previous response's _links.next, for the next page. | | `limit` | integer | no | Page size, default 25. | | `page_id` | string | **yes** | Confluence page id. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `_links` | object | | | `_links.next` | string | | | `results` | array of object | | | `results[].authorId` | string | | | `results[].createdAt` | string (date-time) | | | `results[].message` | string | | | `results[].number` | integer | | Also retrieved by: "show me the edit history of this doc", "who changed this article and when", "what past revisions exist for this write-up". ### `confluence.get_page_version` Get one specific past version of a Confluence page by version number: who made that edit, when, and their message. Use when the user asks about an older revision by number rather than the current content (use get_page for that). Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/api/v2/pages/{{params.page_id}}/versions/{{params.version_number}}`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | | `page_id` | string | **yes** | Confluence page id. | | `version_number` | integer | **yes** | Version number to look up, from list_page_versions. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `authorId` | string | | | `createdAt` | string (date-time) | | | `message` | string | | | `number` | integer | | Also retrieved by: "what did this doc look like three edits ago", "pull up an older revision by number", "show me a specific past copy of this article". ### `confluence.get_current_user` Get the profile of the connected Confluence account itself: accountId, display name, and email. Use to answer "who am I" on Confluence, as opposed to Jira's separate get_current_user. Class `read` (reads only). No scopes beyond the connection's defaults. Calls `GET /ex/confluence/{{params.cloud_id}}/wiki/rest/api/user/current`. Arguments: | Argument | Type | Required | Notes | | --- | --- | --- | --- | | `cloud_id` | string | **yes** | Atlassian Cloud id of the connected Confluence site. | Result fields (the payload is trimmed to these before it reaches the model): | Field | Type | Notes | | --- | --- | --- | | `accountId` | string | | | `accountType` | string | | | `displayName` | string | | | `email` | string | | Also retrieved by: "who am I signed in as here", "what's my confluence identity", "show my own profile on this site".