Freeze release branch v2.4
Cut release/2.4 from main and lock direct commits; require PRs for any fixes.
Release Checklist Board: Ship Software Safely, Every Time
Shipping a release is more than writing code; it's coordinating branch freezes, regression tests, sign-offs, and post-launch monitoring without letting anything slip through the cracks. The Release Checklist Board Kanban Template gives engineering teams a ready-made board with columns for freezing scope, running QA, collecting sign-offs, shipping, and watching the aftermath. Each column comes pre-loaded with specific sample tasks so you're not staring at a blank board, and the whole structure is designed to work with an AI agent through the MCP server, letting you automate task creation and movement as your release progresses.
Template preview
Lock down what's in the release and cut the branch so no new changes sneak in.
Freeze release branch v2.4
Cut release/2.4 from main and lock direct commits; require PRs for any fixes.
Finalize changelog draft
Compile merged PRs since last tag into a draft CHANGELOG.md entry for review.
Confirm feature flags for launch
Review which flags should be enabled/disabled by default in this release.
Run the checks that prove the build is safe to ship.
Run full regression suite
Execute the automated regression suite against the release branch and log failures.
Manual smoke test on macOS app
Install the release build on macOS and walk through core kanban board flows by hand.
Load test REST API endpoints
Hit key REST API routes with expected peak traffic to confirm latency stays within SLA.
Verify MCP server compatibility
Confirm existing MCP tool calls (create_task, move_task, etc.) still behave correctly against the new build.
Get explicit approval from the people who need to say yes before shipping.
Engineering lead sign-off
Get written approval from eng lead that all blocking bugs are resolved or accepted.
Product sign-off on release notes
Product owner reviews and approves the customer-facing release notes for accuracy.
Execute the actual release and make it available to users.
Tag and publish release
Create the git tag, build artifacts, and publish the macOS app update.
Publish release notes
Post the final release notes to the changelog page and notify the community.
Deploy backend REST API changes
Roll out the new REST API version behind the existing deployment pipeline.
Monitor the release for issues in the critical first hours and days.
Monitor error rates for 24 hours
Watch dashboards and MCP server logs for spikes in errors after the release goes live.
Review command palette usage metrics
Check if new command palette shortcuts introduced in this release are being adopted as expected.
Collect early user feedback
Gather feedback from support channels and triage anything urgent into a hotfix task.
Step 1
Freeze the release branch and finalize scope in Scope & Freeze, cutting off new changes.
Step 2
Run automated and manual verification in QA & Verification, logging every failure as its own task.
Step 3
Collect explicit approvals in Sign-off from engineering and product before anything ships.
Step 4
Execute the release in Shipping by tagging, publishing, and deploying to all relevant platforms.
Step 5
Track the release's health in Post-Release Watch, feeding any urgent issues back into a hotfix task.
AI agent usage
An AI agent connected via the MCP server can drive the Release Checklist Board end-to-end: filing pre-release tasks, moving cards through QA gates, tracking sign-offs, and confirming post-release monitoring, all by calling create_task, move_task, and update_task against the correct column names.
## Release Checklist Board
Columns: `Scope & Freeze`, `QA & Verification`, `Sign-off`, `Shipping`, `Post-Release Watch`.
When asked to prep a release, create tasks in `Scope & Freeze` using create_task, then move_task them rightward as gates clear. Never move a task into `Shipping` until it has a completed entry in `Sign-off`. Use get_task before update_task to avoid overwriting checklist notes. Keep total tasks <= 25 (free plan) unless the project is confirmed pro.Learn more about the board-centric workflow in Kanban for AI Agents or open the MCP guide.
The free plan supports 1 project with up to 25 tasks total, which comfortably covers a typical release checklist board like this one with room to spare.
Yes, the columns shown (Scope & Freeze, QA & Verification, Sign-off, Shipping, Post-Release Watch) are a starting point you can rename or reorder to match your team's actual release gates.
Yes, using the MCP server you can connect an AI agent that calls create_task, move_task, and update_task to file and progress release tasks based on your instructions.
You'll need to upgrade to the pro plan ($12/mo or $96/yr), which removes the task cap and the single-project limit so larger releases can be tracked in full.
The board and its MCP integration work the same whether you're managing tasks from the macOS app or through the REST API, so your workflow stays consistent across clients.
Yes, the kanban board supports keyboard shortcuts and a command palette so you can move cards between columns like QA & Verification and Sign-off without reaching for the mouse.
Ready to create this board?
Sign in and use the template to create a project with columns and sample tasks.