Anthropic releases Claude Code dynamic workflows to run parallel subagents for large-scale code migrations

Claude Code just dropped "dynamic workflows" and it's pretty cool.

You type "create a workflow" or turn on "ultracode" in the effort menu and it spins up hundreds of parallel agents that check each other's work.

The unit of work you can hand off jumps from a file to an entire codebase. Migrations, audits, rewrites, framework swaps, stuff you used to plan in sprints now finishes overnight.

The part that got me:....the agents argue with each other before showing you the result. Independent attempts at the same problem, then adversarial agents trying to break the answer. It keeps iterating until they converge. That's how senior engineering teams work. Except this team runs at 3am and never gets tired.

Also if the workflow gets interrupted, it picks up where it left off. That means you can kick off work that runs for days. Not sessions. Days.

Fair warning though: this burns through tokens FAST.

Anthropic says so themselves. But if the task is a codebase migration that would have taken a team 3 months, spending $500 in tokens to do it in a week is the best trade in software.

The ceiling on what one person can build just moved again. Classic.

Going to be playing with this all week.

Pretty cool.

Claude Code just dropped "dynamic workflows" and it's pretty cool.

Someone rewrote 750,000 lines of code in 11 days with it.

You type "create a workflow" or turn on "ultracode" in the effort menu and it spins up hundreds of parallel agents that check each other's work.

The unit of work you can hand off just jumped from a file to an entire codebase. Migrations, audits, rewrites, framework swaps, stuff you used to plan in sprints now finishes overnight.

The part that got me:....the agents argue with each other before showing you the result. Independent attempts at the same problem, then adversarial agents trying to break the answer. It keeps iterating until they converge. That's how senior engineering teams work. Except this team runs at 3am and never gets tired.

Also if the workflow gets interrupted, it picks up where it left off. That means you can kick off work that runs for days. Not sessions. Days.

Fair warning though: this burns through tokens FAST.

Anthropic says so themselves. But if the task is a codebase migration that would have taken a team 3 months, spending $500 in tokens to do it in a week is the best trade in software.

The ceiling on what one person can build just moved again. Classic.

Going to be playing with this all week.

Pretty cool.

This was for an auto research loop on a math problem, so ymmv, but I think this is generally a decent “strategy red-team” prompt:

# Strategy Review

Round ID:

## Result

Choose one:

* `plan_seems_well_scoped` * `coverage_gap_found` * `course_correct_plan` * `full_redirect_recommended` * `meta_planning_round_recommended`

## Red-Team Questions

What inherited assumption could make this search look silly?

Which result would make the last three rounds look misallocated?

If another team beat us by 2x, what search approach would they probably have varied?

Is the current search direction getting stuck in a local minimum?

Are there any radical changes to our approach that we should at least consider?

How would you grade the strategic direction of the last five turns? What were the biggest strategic mis-steps?

If we only had one turn left, what would we do differently? If time and effort were unconstrained what would we do differently?

## Coverage Debt

List search strategies not yet explored but relevant to the stated goal.

## Recommended Next Action

State the next concrete search, survey, verification, or source-review action.