Multi Build — Testing skill for Claude Code
Claude Code skill that builds one new feature with several AI models at once: a spec with testable acceptance criteria and agreed interfaces first, the work split by file across the models, each part.
How to install Multi Build
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Mariomarquezt/multi-build and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Multi Build does
Claude Code skill that builds one new feature with several AI models at once: a spec with testable acceptance criteria and agreed interfaces first, the work split by file across the models, each part reviewed by a different model on a different subscription, one commit.
Alternatives in Testing
- Adversarial Spec — A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until a 514 ★
- Create E2e — Create a Playwright E2E test with explicit success criteria 335 ★
- Toh Framework — Type one sentence, approve once, walk away — Toh Framework installs an AI build department (14 commands, 8 age 94 ★
README
multi-build
**Version 1.0.0** — see [`CHANGELOG.md`](CHANGELOG.md). Your installed version is the `version:` line in `SKILL.md`.
A Claude Code skill that builds **one new feature** with several AI models at once. It writes a spec with acceptance criteria you can test, agrees the interfaces between the parts before anyone codes, splits the work by file evenly across the models, has each builder write the failing test first, has a **different** model on a **different** subscription review every part against the spec, integrates, runs the full suite, and makes one commit.
**The one idea:** parallel building fails at the joins. Each builder makes a private assumption about the seam between its part and the next, and the assumptions disagree. So the seams — signatures, data shapes, routes — are written first, and every brief carries them word for word. Builders fill in behind a contract; they do not invent one.
Sister skills: [`multi-fix`](https://github.com/Mariomarquezt/multi-fix) fixes a list of defects the same way; [`sliced-review`](https://github.com/Mariomarquezt/sliced-review) finds them. All three share one rule: the model that wrote something never checks it, and the check is never on the same bill.
Install
git clone https://github.com/Mariomarquezt/multi-build ~/.claude/skills/multi-build
Recommended alongside it: [`sliced-review`](https://github.com/Mariomarquezt/sliced-review), for the backend wiring notes (`references/runners.md`) this skill points at.
Modes
The number is how many models take part.
| you type | who builds | who reviews | use it when |
|---|---|---|---|
/multi-build 2 |
Sol | Grok | a small feature, one or two files |
/multi-build or /multi-build 3 |
Sol, Grok, Opus — equal shares | each reviews the next, A→B→C→A | the normal case |
/multi-build 5 |
those three + GLM Flash + Composer — equal shares | fixed ring Sol → Opus → Grok → GLM → Composer → Sol | a large feature with many separable parts |
Related Skills
Graftree
Tree-structured, test-first, multi-model solving for hard coding problems. Splits a problem into sub-tasks, wr
Cfn Spec
SPARC Specification phase. Make testable acceptance criteria, edge cases, pre/post conditions, invariants BEFO
Bos Shape
Phase 4 — turn the selected bet into a spec: bounded scope, testable acceptance criteria, flows with full stat
Test Fleet
Test fleet - split a test suite across parallel tester workers with aggregated verdict
Orbits Autopilot
Run the ORBITS flow with the first four gates self-opened and logged, handing back to the person at Test. For
Qikly
Generates a test suite from acceptance criteria, then converges code against it with an agent that never sees
Related Agents
Story Writer
Turns a rough feature idea into a user story with testable acceptance criteria. Output is the first human chec
01 Requirements.Agent
Step 1 of the SDLC pipeline. Turns the Jira story and its acceptance criteria (ACs) into a testable requiremen
Gan Planner
Expand a one-or-two-sentence brief into acceptance criteria, a file list, and a 5-row scoring rubric. First st