Audit legacy auth module for framework coupling
List every file in the legacy auth module and flag hard dependencies on the old framework's session API.
Coordinate large repo migrations with an AI-assisted Kanban board
Migrating a large codebase to a new framework, monorepo, or language version is risky when tracked in scattered docs and chat threads. This template gives engineering teams and their AI agents a single kanban board to plan the migration, run compatibility checks, and manage rollout. Tasks move through columns that mirror a real migration lifecycle, from inventorying legacy modules to verifying production parity. Because the board is exposed over the MCP server and REST API, coding agents can create tasks, update status, and move cards automatically as they complete migration steps, while humans review progress from the command palette or macOS app without losing context.
Template preview
Catalog legacy modules, dependencies, and code paths that must move, and size the migration effort.
Audit legacy auth module for framework coupling
List every file in the legacy auth module and flag hard dependencies on the old framework's session API.
Map database access layer usage across services
Identify all services calling the old ORM directly so we know which need adapter shims.
Tag deprecated internal packages for removal
Search the monorepo for packages marked deprecated and confirm no active service still imports them.
Build shims, adapters, and codemods that let old and new systems run side by side during migration.
Write codemod for import path rewrites
Automate updating import statements from /legacy/* to /core/* across all affected packages.
Build adapter for old REST API contract
Create a compatibility layer so existing REST API clients keep working during the transition.
Add feature flag for dual-write to new schema
Implement a flag so writes go to both old and new database schemas until cutover is verified.
Backport bug fixes into legacy branch
Ensure critical fixes land in both the legacy and migrated branches until rollout completes.
Validate migrated code paths with automated checks before they reach production traffic.
Run full regression suite against migrated services
Execute the existing test suite against the migrated module and log any failures for triage.
Diff API responses between old and new endpoints
Compare response payloads from old and new REST API implementations for parity on key routes.
Load test migrated data layer
Confirm the new data access layer meets latency targets under production-like load.
Ship migrated components to production behind flags and confirm stability before cleanup.
Enable migrated checkout flow for 5% of traffic
Gradually ramp traffic to the migrated checkout service and watch error rates.
Set up alerting on migrated service error budget
Configure alerts so any regression in the migrated path is caught before full rollout.
Remove legacy code path after full cutover
Once rollout is stable at 100%, delete the old implementation and update documentation.
Step 1
Create one project for the migration and use the MCP server or REST API to bulk-import inventoried modules as tasks in Inventory & Scope.
Step 2
As engineers or agents identify compatibility gaps, create tasks in Compatibility Work and move related inventory tasks forward once shims exist.
Step 3
Move tasks into Testing & Checks when a codemod or adapter is ready, attaching regression and diff results to each task.
Step 4
Once checks pass, move tasks to Rollout & Monitoring and track flag ramp-up percentages directly in task details.
Step 5
Use keyboard shortcuts and the command palette to triage stalled tasks daily, keeping the board under the 25-task free plan limit or upgrading to pro for larger migrations.
AI agent usage
An AI coding agent can drive this board end-to-end: scanning the repo to populate Inventory & Scope, opening Compatibility Work tasks when it detects breaking changes, and moving tasks through Testing & Checks and Rollout & Monitoring as it runs test suites and ramps feature flags via the MCP server tools.
## Migration Board Rules
When operating on the Codebase Migration Board:
1. New legacy modules discovered during repo scans go into 'Inventory & Scope' via create_task.
2. When a breaking change requires a shim, create a task in 'Compatibility Work' and link it to the source task.
3. Only move a task to 'Testing & Checks' after codemods or adapters are committed.
4. Only move a task to 'Rollout & Monitoring' after regression and diff checks pass.
5. Never exceed the plan's task limit; check list_tasks count before create_task.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. This template ships with 14 sample tasks across 4 columns, leaving room to add your own before you'd need to upgrade to pro.
Yes. Both humans and agents can read and update the board through the REST API or MCP server, so an agent can create or move tasks while engineers review from the macOS app.
You can upgrade to the pro plan for $12/mo or $96/yr, which removes the single-project and 25-task limits so you can track a larger migration in full detail.
Yes, columns like Inventory & Scope, Compatibility Work, Testing & Checks, and Rollout & Monitoring are a starting point you can rename to match your team's terminology.
Define clear rules in your AGENTS.md file, such as only moving tasks to Testing & Checks once code is committed, so agents using get_task and update_task follow consistent criteria.
Yes, you can use keyboard shortcuts and the command palette to quickly jump between tasks and columns during daily migration standups.
Ready to create this board?
Sign in and use the template to create a project with columns and sample tasks.