Clear a backlog of small tasks overnight
Point a team at thirty low-risk tickets and wake up to thirty diffs waiting for review.
Small tasks are individually trivial and collectively fatal. Nobody wants to spend Thursday on copy tweaks, dependency bumps and missing null checks — so they rot in the backlog for a year.
The boring 80% is already written and reviewed once. Your job shrinks to judgement calls.
The team to build first
Copy this composition, then add builders only once cards start waiting in assigned instead of review.
- LeadReads the list, sizes each task, assigns in dependency order
- Builder ×3Work tickets in parallel, each in its own workspace
- ReviewerFirst-pass review, bounces anything that touches shared code
Bump deps to latest minor
Add rate limit to /api/sync
Extract auth into a module
Write tests for parser
Fix null deref in session
Migrate config loader
How the run goes
- 01
Import the list
Paste the tickets or attach the file. Each becomes a card in the backlog column with its own context.
- 02
Set the guardrails
Cap the monthly token budget, mark which paths are off limits, and require review before done.
- 03
Let it run
Builders claim cards in parallel. The reviewer checks each diff and sends weak ones back with a comment.
- 04
Review in the morning
You see a review column, not a wall of logs. Accept the good ones, comment on the rest.
- 02
Kanban board
Five columns that move on their own. Tasks change status the moment an agent picks them up.
- 01
Agent teams
Build a team with real roles. Agents work in parallel, talk to each other, and hand off across teams.
- 03
Code review
A diff per task with accept, reject and comment — plus a built-in editor with Git support.
- 04
Token analytics and budgets
See exactly where tokens go by team, agent, task and model. Set caps and get warned at 80%.
- AnswerWhat stops agents from burning money overnight?Set a monthly ceiling in tokens or currency. You get an alert at 80% and again at 100%, and scheduled runs can hard-stop at the cap rather than continuing. Spend is broken down per team, agent, task and model so you can see what is expensive.
- Use caseShip a product as one person with a team of agentsStart in solo mode, promote to a team when the work outgrows one agent, and keep the spend capped.
- Use caseUse agents without sending your repo to a vendorCoordination stays on your machine, and you choose which model — including fully local ones — sees any code.
Run this workflow tonight.
Install, copy the team above, and give it the objective. Free, no account.