Jira and Confluence,one shortcut away
A ticket in Jira, the decision on a Confluence page: the same story split across two tabs, always a little out of sync. Khint puts both behind one palette: pull an issue, a sprint, or a page in as context, then create, comment, or publish back, without leaving the app you're in.
The tracking and the doc
never quite agree
Jira holds the state of the work; Confluence holds the reasoning behind it. Keeping them consistent means bouncing between tabs, copying a decision from a page into a comment, updating a status you already wrote up somewhere else. Khint closes that gap with its three pillars (Agents, Capture, and Memory) wired behind one palette, so your Atlassian workspace is something you drive from a keyboard shortcut instead of a browser you keep tab-switching in.
One shortcut, both surfaces, both directions
Pull the sprint and the page in as context; write the ticket and the doc back out.
Connect Atlassian once
On the MCP page, open the Atlassian card and enter your email, an API token, and your cloud domain. Those three values land in your operating system's keychain. The connection is local-only and PAT-based, so nothing about your Jira or Confluence routes through Khint's servers. From now on, both sit behind one shortcut.
Pull the sprint and the issue into context
You're mid-sprint and about to work a ticket. Open the palette with Cmd+Shift+K and run Active sprint, then Pull issue for the one you're on. With a Memorysession active, each pull is logged to it, so the sprint and the issue become part of your compacted context. The thread of what you're working travels with you.
Bring in the Confluence page that documents it
The decision behind the ticket lives on a Confluence page. Run Pull page to read it into the same session. Now Khint holds both halves of the story at once (the tracking in Jira and the documentation in Confluence), so nothing you do next starts from a blank prompt or a half-remembered spec.
Write back to both, and keep them in sync
A decision gets made, so you file it. Run Create page to publish the write-up to Confluence, then Add comment or Change status on the Jira issue. The Create flow runs your text through a saved Action first, and because the session context rides along, the doc and the ticket update reflect the same source of truth instead of drifting apart.
Or chain the whole round trip
Turn the recurring version into a workflow. A Khint chain runs up to five steps, and a step can be an integration write as well as an AI action, so one keystroke can shape raw notes into a ticket, file it to Jira, then create the matching Confluence page. The running text flows from step to step, so a trailing step can paste the created issue's link. Turn meeting notes into a Confluence page →
Why not just keep both tabs open?
You can, but two tabs is exactly what lets the ticket and the doc drift, because each edit happens in isolation with no shared thread between them. Khint carries that thread: pulls from Jira and Confluence land in the same Memory session, and every write is shaped from that one context. The Atlassian calls go straight from your Mac to your cloud with your own token. Reads and writes are local-only and PAT-based, credentials sit in the keychain, and each write is one item per run, so nothing is filed silently. Want the whole PO/BA stack behind the palette? See all the integrations →
Common questions
Can Khint drive both Jira and Confluence at once?
Yes: one Atlassian connection powers both. Once you connect on the MCP page, the palette behind Cmd+Shift+K gains a JIRA section (pull issue, active sprint, search, create issue, add comment, change status) and a CONFLUENCE section (pull page, create page). So the ticket you are tracking and the doc that explains it sit behind the same shortcut, and you never open a browser tab to jump between them.
Can I pull an existing issue or Confluence page in as context?
Yes, and that is the point. The Jira reads (pull issue, active sprint, and search) and the Confluence read (pull page) bring existing Atlassian content into Khint. When a Memory session is active, each pull is logged to it, so the issue you are working and the page that documents it become part of your compacted session context. Your next AI action then already knows both, instead of starting cold.
How does Khint keep a ticket and its Confluence doc consistent?
Pull the issue and its related page into an active Memory session first. Because that compacted context rides along on every AI action, a Create page or Add comment run afterwards is shaped from the same thread. The decision you file to Confluence and the status you drop on the Jira issue reflect the same source of truth, so the tracking and the documentation do not drift apart.
How do I connect Atlassian, and where do my credentials live?
On the MCP page, open the Atlassian card and enter your email, an API token, and your cloud domain. Those three values are stored in your operating system's keychain, not on Khint's servers. The connection is local-only and PAT-based. You can hide any palette row you don't use from the same card, so your JIRA and CONFLUENCE sections show only the actions you actually run.
Keep Jira and Confluence in step
Free with 10 AI actions and 5 captures per day. No credit card. Connect Atlassian once, then run both from one shortcut.