Raise test coverage on a project nobody tested
One agent per module, tests written in parallel, every suite run in the built-in terminal before review.
Writing tests is the most delegable work in software and the least fun. Coverage targets get set, ignored, and quietly dropped.
Coverage moves in days, and the tests are ones you would have written.
The team to build first
Copy this composition, then add builders only once cards start waiting in assigned instead of review.
- LeadRanks modules by risk and coverage gap
- Test writer ×3One module each, tests written in parallel
- ReviewerRejects tests that assert nothing meaningful
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
Find the gaps
Run the coverage tool in the built-in terminal and turn the report into cards.
- 02
One module per agent
Separate workspaces mean three agents can write tests at once without touching the same files.
- 03
Green before review
Each agent runs the suite itself. A card only reaches review when the tests actually pass.
- 04
Guard the quality
The reviewer flags assertions that only test the mock. Comments go back as revision instructions.
- 01
Agent teams
Build a team with real roles. Agents work in parallel, talk to each other, and hand off across teams.
- 06
Terminal workspace
A real terminal in the app, with team shells, local shells, history and autocomplete.
- 03
Code review
A diff per task with accept, reject and comment — plus a built-in editor with Git support.
- 02
Kanban board
Five columns that move on their own. Tasks change status the moment an agent picks them up.
- 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.
- AnswerHow many agents should I run at once?Start with three: a lead, a builder and a reviewer. Add builders only when the board shows cards waiting in assigned rather than in review. Most people over-provision builders and end up bottlenecked on review instead.
- Use caseRefactor a legacy codebase without breaking itSplit the refactor across agents, block risky work behind tests, and review every diff before it lands.
Run this workflow tonight.
Install, copy the team above, and give it the objective. Free, no account.