Read and write Confluence Cloud pages, spaces, blog posts, comments and tasks.
Confluence ships in the w6w first-party pack. It declares 22 actions, 3 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.confluenceConfluence Cloud is Atlassian’s team wiki, and this app reads and writes its pages, spaces, blog posts, comments, and inline tasks — everything a workflow needs to publish documentation, keep a knowledge base current, or pull content into another system. Pages can be created, updated, listed by space or title, and searched with Confluence’s own query language; spaces can be created and browsed, including as a tree of child pages; and blog posts can be published into a space the same way.
Comments and tasks round it out: the discussion thread at the bottom of a page can be read and posted to, and a workflow can list the inline action items assigned across a space, filterable by status or assignee. Content search covers the single most useful thing a workflow does with a wiki — finding what mentions something and has changed recently — even where Confluence’s current API doesn’t otherwise offer a search endpoint of its own.
Three routes to the same 22 actions. The Workflow tab is generated from Confluence's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
comment-create Add a footer comment to a page or blog post, or reply to another comment.
space-page-list List the pages in one space, optionally only the root-level ones.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Confluence's real ids.
{
"manifestVersion": "2",
"name": "confluence-example",
"steps": [
{
"id": "attachment-get",
"uses": {
"app": "io.w6w.confluence",
"action": "attachment-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"attachmentId": "<attachmentId>"
}
}
]
}attachment-get blogpost-create blogpost-get blogpost-list comment-create +17 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Confluence, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Confluence. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "attachment-get",
payload: {
attachmentId: "<value>",
// version: "<value>",
// includeLabels: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action attachment-get --payload '{"attachmentId":"<value>"}' Give an AI agent Confluence — without giving it Confluence's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.confluence#attachment-get",
"input": {
"attachmentId": "<attachmentId>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Confluence connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Confluence's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Confluence itself is MIT, and the runtime that executes it is source-available (FSL).
Confluence declares its own checks, so its health is a property of the app rather than something the host guesses at.
Atlassian Statuspage rollup for confluence.status.atlassian.com, with per-component detail. Unauthenticated and unsigned.
Unauthenticated `GET /status` against this connection's Atlassian site — proves the site exists and is serving.